Historic data from Union Carbide's drilling and metallurgical testwork have added geological and processing support to OD6 Metals’ Quinn Fluorspar Project, positioning the Nevada asset as a developing US critical minerals opportunity. Samso News | ASX: OD6 | Fluorspar | Nevada | Acidspar | Critical Minerals | Union Carbide OD6 Metals Limited (ASX: OD6) has added another layer of technical foundation to the Quinn Fluorspar Project in Nevada, USA (Figure 4), with historic Union Carbide drill data confirming high-grade, near-surface fluorspar mineralisation at Horseshoe. The company had recently acquired a rare dataset of geological records from Union Carbide Corporation. On analysing this historic data, OD6 delineated historic grades up to 93.7% CaF₂, - interpreting the system as open up-dip, down-dip and at depth. Together with historical metallurgical testwork indicating potential acidspar production . Quinn is now developing into a more defined US critical minerals opportunity pending modern validation. Samso Summary Table 1: OD6 Metals Summary Item Details Company OD6 Metals Limited ASX Code OD6 Project Quinn Fluorspar Project Location Nevada, USA Commodity Fluorspar / fluorite Main Prospects Referenced Horseshoe, Mammoth, Big Jim Key Drilling Update Historic Union Carbide drilling confirmed high-grade, near-surface fluorspar mineralisation at Horseshoe Key Metallurgical Update Historic testwork indicated potential to produce acidspar through conventional grinding and flotation Best Historic Drill Grade Peak result of 93.7% CaF₂ over 1.2m in H1 Key Historic Metallurgical Result Mammoth testwork achieved 97.83% CaF₂ at 70% recovery and 95.0% CaF₂ at 82.7% recovery Product Relevance Potential production of metspar and acidspar, subject to modern confirmation Next Work Programs Modern assays, geological modelling, channel and rock-chip sampling, soil geochemistry, drill target definition, TOMRA optical ore sorting and flotation testwork Provenance of the Dataset The Union Carbide work at Quinn was conducted across two seasons. In 1957, Union Carbide Nuclear Company and The Galigher Company in Salt Lake City completed preliminary metallurgical amenability testing on samples from Mammoth and Horseshoe. In 1958, Union Carbide returned to drill four diamond holes at the Horseshoe deposit, supported by a summary report prepared by H.M. Wharton dated 27 August 1958. A second metallurgical program was scheduled to commence at The Galigher Company in July 1958, with new samples already despatched (Figure 1). Figure 1: Example Drill map of UCC Horseshoe (AKA Higrade) deposits, Nye County, Nevada (Source: OD6 ASX Announcement May 7, 2026) The 1958 follow-up program did not proceed. A local prospector disputed Union Carbide's claims and denied the Company access to the site and to a water supply, and the program was discontinued. The dataset has remained unpublished for the intervening six and a half decades. OD6 acquired the full archive of Union Carbide records relating to Quinn, on terms disclosed in the 7 May 2026 ASX announcement. The acquired material includes high-quality scans of cross-sections, maps, geological logs, original assay sheets, and the underlying Galigher and Union Carbide Nuclear Company metallurgical reports. The Competent Persons engaged by the Company have reviewed the underlying records and have concluded that the reported head grades are consistent with grades observed in recent OD6 field sampling, and that the flotation reagents and processing techniques described remain broadly consistent with conventional fluorspar processing methods used today. Historic Drilling at Horseshoe Union Carbide Corporation’s drill program consisted of four shallow drill holes, H1 to H4, with an average depth of approximately 38 metres (Figure 2). OD6 reported that the drilling confirmed high-grade fluorspar mineralisation from surface and intersected significant widths of CaF₂ mineralisation. Table 2: Reported Historic Drill Intercepts Hole Reported Intercept H4 14.3m @ 70.9% CaF₂ from surface H2 14.0m @ 59.9% CaF₂ from 19.5m H1 25.8m @ 46.2% CaF₂ from 0.9m H3 6.4m @ 46.2% CaF₂ from 22m Within H1, OD6 reported: Interval Result 12.8m from 0.9m 60.9% CaF₂, including a peak result of 93.7% CaF₂ over 1.2m 3.4m from 16.8m 58.8% CaF₂, including a peak result of 82.0% CaF₂ over 1.2m 3.5m from 22.3m 43.7% CaF₂, including a peak result of 66.1% CaF₂ over 0.3m The historic drilling targeted the southern corner of the system and did not test the high-grade core later exposed in open-pit mining. The assaying only tested bulk fluorspar intercept zones and that multiple sections of core were not assayed despite being logged with fractured breccia and/or vein material that may be fluorspar-bearing. OD6 interprets the mineralisation at Horseshoe as a fault-controlled replacement system with an exposed footprint of approximately 3,000 square metres. The company states that the system remains open up-dip, down-dip and at depth. Figure 2: Historic Drilling Results (Source: OD6 ASX Announcement: May14, 2026) Management Commentary “Historic drilling at Horseshoe indicates a rare combination of exceptional high grade and significant widths, with results well above typical global fluorspar benchmarks. Importantly, these intercepts were returned from the margin of the system, not the high-grade core later exposed in open pit mining (and reported in channel sampling results by the Company 15 April 2026) and were notably only selectively assayed. This indicates the system has the potential to be both higher grade and larger than intercepted in historic drilling. – Brett Hazeldon Managing Director and CEO, OD6 Metals Ltd Historic Metallurgical Testwork In addition, Union Carbide also completed metallurgical testwork on fluorspar mineralisation from Mammoth and Horseshoe. The Mammoth sample had a head grade of approximately 28.29% CaF₂, while the Horseshoe sample had a head grade of approximately 58.57% CaF₂. Historical Union Carbide and Galigher Company testwork indicated concentrate grades of 95.0% CaF₂ at 82.7% recovery and 97.83% CaF₂ at 70% recovery from Mammoth mineralisation. These historical results suggest that Quinn mineralisation may be capable of producing acidspar using conventional grinding and flotation methods, subject to confirmation through modern metallurgical testwork. OD6 also notes that the historical testwork was preliminary and unoptimised. The historical Union Carbide metallurgical testwork provides encouraging evidence that fluorspar mineralisation at Quinn may be amenable to conventional processing techniques. Importantly, the historic results suggest the potential to upgrade both low-grade and high-grade mineralisation to produce commercial fluorspar concentrates, including acid-grade product, subject to confirmation through modern metallurgical testwork. Acidspar and Metspar Product Context Fluorspar, also known as fluorite, has the chemical formula CaF₂ and contains 48.9% fluorine by weight. OD6 describes fluorspar as the main mineralogical source of fluorine. OD6 identifies two main product categories: Table 3: Difference between Metaspar and Acidspar Product Description Metspar A mineral product containing greater than 60% CaF₂, used in the steel industry as a fluxing agent Acidspar An upgraded mineral product containing greater than 97% CaF₂, used for hydrofluoric acid production OD6 states that acidspar is used in the chemical industry for hydrofluoric acid production, with downstream applications in nuclear fuels, battery technologies, solar panels, defence technologies and AI chip manufacturing. Next Phase Metallurgical Program OD6 is advancing a staged metallurgical testwork program to validate and optimise processing flowsheets. Recent OD6 field programs have collected extensive metallurgical samples across multiple Quinn prospects for modern testwork. The company plans to send samples for optical ore sorting testwork with TOMRA in Germany. The stated objectives are to upgrade feed grade prior to processing, reduce processing costs and plant size, and assess reject/waste separation efficiency. OD6 expects TOMRA testwork to commence this quarter, with results expected in quarter three. OD6 also plans additional flotation testwork covering flotation optimisation, dense media separation, grind size and reagent testing, and product specification validation for acidspar versus metspar. The company expects flotation testwork to commence this quarter, with results anticipated in the second half of the year. OD6 also intends to apply for bulk sample permits across multiple areas to support advanced metallurgical testwork, representative pilot-scale testing, flowsheet development and potential offtake samples. With extensive representative samples now collected from across the project, OD6 is preparing to commence modern metallurgical programs, including optical ore sorting with TOMRA and flotation optimisation with specialist laboratories. These programs are designed to validate the historical results and support the development of Quinn as a potential domestic US source of high-grade fluorspar. Figure 3: : Quinn Fluorspar Project with, deposit locations, background geology and alteration map (Source: OD6 ASX Announcement May 20, 2026) Due Diligence and Next Steps OD6 has outlined the following work programs as part of its due diligence and validation process at Quinn: Digitise scanned paper logs and cross-sections into a geological model. Receive and interpret assay results. Expand systematic channel and rock-chip sampling. Validate and replicate historic high-grade results. Undertake detailed geological and structural mapping. Complete soil geochemistry programs. Identify and prioritise drill targets. Initiate permitting for maiden drilling. Progress metallurgical testwork planning. The system remains open in all directions and untested at depth, which in conjunction with recent fieldwork confirming extensive mineralisation, provides significant exploration upside and a clear pathway for targeted drilling to define a large scale fluorspar system. We also look forward to results from our recent mapping and sampling programs, which will further define the scale and potential of Horseshoe The Fluoride Market - The Basics Fluorspar (the commercial name for the mineral fluorite, CaF₂) is the world's primary source of fluorine. The market is roughly 8-9 million tonnes per year and valued at around US$2.1-2.8 billion in 2025-2026, with most forecasters expecting low-to-mid single-digit CAGR growth through to 2032-2035. It is split into two main commercial grades: acid-grade (acidspar, ≥97% CaF₂) which feeds the chemical industry, and metallurgical-grade (metspar) which goes into steel and aluminium production. The headline structural feature of this market is extreme geographic concentration on the supply side (Figure 6). Figure 6: Global players in the Fluorspar market. China, Mexico and Mongolia together account for ~84% of global supply. China's position is structurally important because it is also the largest consumer — Asia Pacific accounts for roughly 74% of global volume — meaning Chinese industrial activity and export policy effectively set the global price. The flip side of this concentration is that fluorspar is now formally listed as a critical mineral by the United States, EU, China, Canada, Japan and Australia, and Western governments are actively supporting new supply (the Lost Sheep mine in Utah, the St Lawrence mine restart in Canada, plus projects in Australia, Germany and Kenya). On the demand side, the market is dominated by chemical applications rather than the traditional metallurgical uses most people associate with the mineral. Where fluorspar goes — grade split and downstream end-use The outlook narrative (see Figure 7) — and the reason fluorspar is getting attention now — is the shift in chemical demand toward energy-transition end-uses. The traditional refrigerant market is actually being phased down (the US AIM Act has cut HFC production allowances to 40% below the historic baseline), but this is being more than offset by three rising demand vectors: Lithium-ion batteries. A Li-ion EV battery uses 5-10x more fluorspar (by mass) than lithium — fluorine compounds appear in the cathode binder (PVDF), the electrolyte salt (LiPF₆), and separator coatings. Benchmark Mineral Intelligence estimates the battery segment alone will pull more than 1.6 Mt of fluorspar annually by 2030, growing at over 20% CAGR. Semiconductors. Ultra-high-purity hydrogen fluoride is essential for etching silicon wafers, with data centre build-out for AI lifting demand. Decarbonisation chemistry. Hydrogen fuel cells, green refrigerants, and fluoropolymer membranes all consume acidspar-derived HF. Figure 7: The distribution of its uses. The US Department of Energy has projected that under current trajectories, fluorspar demand will exceed current supply by 40-70% by 2035. That gap, combined with the geographic concentration shown in the first chart (Figure 6), China's own declining reserves, and the critical mineral designations is what is bringing new Western producers and developers into the conversation. It is also worth noting that fluorspar is rarely produced as a by-product; it depends on dedicated mining operations, which makes the supply side less responsive to sudden demand spikes than commodities like cobalt or molybdenum. Samso Concluding Comments The OD6 story in the last few months has been transforming and the speed in which data has been released is comforting for shareholders. As you can see in Figure 5 below, the resurrection of the share price is promising but I do feel that there is a certain case of fatigue coming onto the screen. The recent frequency of releases may be counter-intuitive to the journey. Figure 5: The share price chart for OD6 Metals Limited as of 26th May 2026. (source: Commsec) One cannot choose the fortunes and misfortunes but the good part of the story is there is urgency and management is definitely trying to make this work. Personally, this is an exciting project as the data slowly unearth the prospectivity of the situation in Nevada. The market for Fluoride or Fluorspar is not something the general public comes across but that is a good thing for OD6 Metals allowing a "educational" educational theme. What I like about the fluoride industry is that there does not appear to be any new upcoming stories that would create problems for OD6 Metals. Most importantly, the grade that are being talked a bout for OD6 Metals appear to be in the Tier-1 category, so lets see how that plays out over time. About OD6 Metals Limited OD6 Metals Ltd (ASX: OD6) is an Australian public company pursuing exploration and development opportunities across the critical minerals sector, with a portfolio spanning fluorspar, rare earth elements, and copper. Its flagship rare earth asset is the Splinter Rock Project in Western Australia's Esperance-Goldfields region, which hosts one of Australia's largest clay-hosted rare earth deposits with an Indicated Resource of 119Mt at 1,632ppm TREO and an Inferred Resource of 563Mt at 1,275ppm TREO. In fluorspar, the company holds an option to acquire the Quinn Fluorspar Project located approximately 220km north of Las Vegas, Nevada — a project with documented high-grade mineralisation across multiple systems, including Mammoth, Horseshoe, and now the rediscovered Big Jim lode, each exhibiting grades well above the economic threshold for fluorspar development. The company also holds the Gulf Creek Copper-Zinc VMS Project near Barraba in New South Wales (Figure 4). Figure 4: Location and neighbourhood of Quinn Flourspar Project in Nevada Previous Samso News Coverage Samso has followed OD6 Metals Ltd across multiple ASX releases. The following represents our prior published coverage of the company: May 19, 2026 - April 05, 2026 - April 17, 2026 - March 18, 2026 - December 19, 2025- November 08, 2025- September 10, 2025- August 24, 2025 - August 13, 2025 August 22, 2023 Cover photo: Mexico-fluorite (3), Romaine. CC0, via Wikimedia Commons (https://commons.wikimedia.org/wiki/File:Mexico-fluorite_(3).jpg) The Samso Way – Seek the Research Here at Samso, we pride ourselves on delivering content for investors that is independent and informed by over three decades of experience in the industry. Our content is well-researched and is only created if I see merit in discussing the company's story. Our mission is simple: cut through the noise and spotlight what matters—genuine stories, grounded insights, and real opportunity. Our content is well-researched and is only created if the team sees merit in discussing the company or concept. Investors can explore our three core platforms: Coffee with Samso Samso Insights Samso News There may be numerous paths to success in investing, but the common thread among successful individuals is that they remain committed to making informed decisions. Equip yourself with the right knowledge and tools, and you will be well on your way to achieving your financial goals. Most importantly, investors need to be absolutely diligent in understanding their own risk-reward tolerance and capabilities. Never bite off more than you can chew. As they say, Rome wasn’t built in a day, and the Great Wall stood because it took centuries to complete. The Samso Philosophy: Stay curious. Stay sharp. And remember—digging deeper always uncovers the real value. In Life, there is no such thing as a Free Lunch. Never bite off more than you can chew is my parting comment. Happy Investing, and the only four-letter word you need to know is DYOR. To support our independent nature of our work, please head over to our Support Page and give us a helping hand in any of the ways listed. This is a new initiative for the Samso Platform, and it was always the concept of Samso when we started this journey in 2018. Disclaimer The information or opinions provided herein do not constitute investment advice, an offer, or solicitation to subscribe for, purchase, or sell the investment product(s) mentioned herein. It does not take into consideration, nor have any regard to your specific investment objectives, financial situation, risk profile, tax position and particular, or unique needs and constraints. Read full Disclaimer. Share to Grow: Your Bonus Samso has just released an eBook: How to Add Value to your Share Portfolio |A lesson on geological models sought by mining companies that gives insight and an understanding of which portfolios are better - and potentially more lucrative – investments. Click here to download this eBook.| Download eBook If you find this article informative and useful, please help me share the information. I try to write about topics that are interesting and have the potential to be of investment value. It is not easy to find stories that fit those parameters. If you or your organisation sees the benefit of what Samso is trying to achieve and has a need to share your journey, please contact me at noel.ong@samso.com.au. About Samso Samso is a trusted platform that equips dedicated investors with up-to-date industry knowledge and insigh0ts from top CEOs and thought leaders. By staying informed on business advancements and market trends, investors can enhance their financial decisions through a combination of expert guidance and their own research. Samso News | www.samso.com.au | An Investor Lens on ASX-Listed Companies

A gold-copper explorer betting on the Paterson Province pedigree — and the team that helped find Hemi 01 / The 60-Second Pitch Aventine Resources Ltd (proposed ASX: ARE) is a Western Australian gold-copper explorer coming to the ASX with a $15m (minimum) to $20m (maximum) raise at $0.20 per share. The company is asking the market to back four projects totalling roughly 1,780 km² in the Paterson Province and the East Pilbara — two of the more credentialed exploration addresses in the country. The pitch is, first and foremost, about location. Aventine’s flagship Paterson Project sits north-northwest of Greatland Resources’ Havieron gold-copper deposit (4.1Moz Au) and northeast of the Telfer mine (3.2Moz Au) (Figure 1). The broader district hosts Rio Tinto’s Winu copper-gold discovery, the Nifty copper mine, Antipa’s Minyari Dome, and — over in the Pilbara — De Grey’s 11Moz Hemi gold system. This is not frontier dirt nobody has heard of. It is ground that the majors have circled for forty years. Figure 1: Map of Aventine Resources project tenements across the Paterson Province and East Pilbara, Western Australia, near the Telfer and Havieron gold-copper deposits (Source: IPO Prospectus) The notable name on the register is Greatland Resources (AIM; ASX: GGP). Greatland is vending much of Aventine’s tenure into the float, and hence they will be the company’s largest shareholder at around 11% on listing, and is appointing a nominee director (Rowan Krasnoff) after admission. Greatland owns and operates the two anchor assets of the modern Paterson — Telfer and Havieron — so its decision to seed Aventine and stay on the register is the closest thing this deal has to a stamp of approval. The second pillar is people. Technical Director Allan Kneeshaw is a 30-year exploration geologist who, most recently at De Grey Mining, played a key role in the discovery of the 11Moz Hemi gold deposit. Exploration Manager Rod King also comes out of the De Grey / Northern Star camp via a decade at Teck. Managing Director Benjamin Dunn brings the capital-markets side — two decades across Citigroup, JP Morgan and CLSA. On paper, this is a team that has actually found something before. What you are not buying is a resource. There is no JORC Mineral Resource, no Ore Reserve and no Exploration Target anywhere in the portfolio. There are encouraging historical drill intercepts — and that is exactly the point. Aventine is a drilling story, and the budget is built to drill. 02 / Aventine Resources IPO Snapshot Everything you need to know about the Aventine Resources IPO in one table: Table 1 — IPO Snapshot Item Detail Company Aventine Resources Ltd (ACN 686 650 297) Proposed ASX Code ARE Offer Price $0.20 per share Raise (min / max) $15.0m / $20.0m (75m / 100m shares) Indicative Market Cap (min / max) ~$27.95m / ~$33.51m (undiluted, at offer price) Existing Shares on Issue 31,652,222 Shares on Issue at Admission 139,767,376 (min) – 167,545,154 (max) Free Float ~62% (not less than 40%) Lead Manager Bell Potter Securities Limited Cornerstone / Strategic Greatland Resources (AIM; ASX: GGP) – ~11.1–11.4%; nominee director post-listing Lead Manager Options 3,000,000 unquoted @ $0.30, 3-yr expiry Board Options 7,000,000 unquoted @ $0.25 and $0.35, 3-yr expiry Performance Rights 8,884,168 (plus a separate Mt Cecelia milestone right) Lodgement / Open / Close / Quotation 22 May / 1 June / 22 June / 7 July 2026 (indicative) Min Application 10,000 shares ($2,000), then 2,500-share ($500) increments Underwritten? No Eligible Jurisdictions Australia (with limited offers to certain investors in permitted jurisdictions) 03 / Capital Structure & Dilution This is the number to sit with before anything else: existing shareholders go from owning 100% of the company to approximately 19% on Maximum Subscription (around 23% on Minimum). The number of shares on issue increases from 31.65 million to 167.5 million on maximum subscription — a ~429% increase. On a fully diluted basis (all options and performance rights in, plus the Mt Cecelia milestone shares), there will be roughly 191 million securities in issue at maximum. That is heavy, and heavier than you see on many small-cap floats. It is the direct consequence of a thin pre-IPO share count (a young company) raising a relatively large $15–20m and issuing a big slug of consideration shares to the vendors. New money and vendors together own roughly 75–80% of the company on day one. Existing holders are very much along for the ride. Table 2 — Capital Structure & Dilution Security class Min sub. (shares) % Max sub. (shares) % Existing shares 31,652,222 22.6% 31,652,222 18.9% IPO Offer shares 75,000,000 53.7% 100,000,000 59.7% Vendor shares 33,115,154 23.7% 35,892,932 21.4% Total shares at Admission 139,767,376 100.0% 167,545,154 100.0% Options on issue 10,000,000 — 10,000,000 — Performance Rights 8,884,168 — 8,884,168 — Fully diluted total 158,651,544 — 186,429,322 — Fully diluted total excludes the Mt Cecelia milestone right (up to a further 5,000,000 shares). The options and rights stack There are 10,000,000 options on issue at admission: — 3,000,000 Lead Manager Options @ $0.30, expiring 3 years from issue — 7,000,000 Board Options @ $0.25 and $0.35, expiring 3 years from issue Plus 8,884,168 Board Performance Rights issued to directors and management, with vesting terms set out in the prospectus. Separately, the Mt Cecelia acquisition carries a milestone right that converts to up to 5,000,000 shares if a JORC resource of at least 500,000 ounces of gold is defined at that project within five years. Two things to like here. First, the board options strike at and above the IPO price ($0.25 and $0.35), so management only makes money on them if the stock works for everyone. Second, the milestone right is genuinely performance-linked — it only triggers on a half-million-ounce discovery, which is the kind of outcome every shareholder would happily be diluted for. Escrow Roughly 52.7 million shares will be classified as restricted by the ASX — about 23.8m shares (plus most options and rights) under 24-month escrow, and about 28.9m shares under 12-month escrow. None of the IPO shares are escrowed. The free float is expected to be approximately 62%. That is a healthy free float, but the escrow means a meaningful chunk of the register is locked up for the first year or two — worth keeping in mind for liquidity. 04 / Use of Funds Including the $2.11m of cash already on the balance sheet, Aventine will have roughly $17.1m (min) to $22.1m (max) to deploy. Here is where it goes over the two years following admission: Table 3 — Use of Funds (2 years) Use of funds Min sub. ($) % Max sub. ($) % Exploration expenditure 11,981,274 70.0% 16,692,169 75.5% Feasibility expenditure 400,000 2.3% 400,000 1.8% Directors’ & management fees 1,991,360 11.6% 1,991,360 9.0% Costs of the Offer 1,382,511 8.1% 1,689,820 7.6% Working capital 1,353,996 7.9% 1,335,792 6.0% Total 17,109,141 100.0% 22,109,141 100.0% This is a lean, exploration-weighted budget, and that is a genuine positive. Putting 70–75% of available funds straight into the ground is at the upper end of what you see on a float of this size — a lot of explorers come to market spending a third of the raise on overheads and the offer. Aventine isn’t. The one line that may appear to be of note is costs of the offer, which work out to roughly 8–9% of the gross amount raised (higher at the minimum end). That is not outrageous for a non-underwritten small-cap IPO, but it sits at the firmer end of the range, and the Lead Manager also collects 3 million options on top of the cash fee. Directors’ and management fees of ~$2.0m over two years are reasonable for a five-person board plus management, and the percentage falls as the raise grows. 05 / The Project Portfolio Aventine comes to market with four projects across the Paterson Orogen and the East Pilbara. The headline acreage is real, but read the granted-versus-application column carefully — a chunk of the Paterson South ground is still in application. Table 4 — Project Portfolio Project Area (km²) Tenements Targets Lever / Analogue Paterson 676 5 granted Au, Cu NNW of Havieron, NE of Telfer; Black Hills Dome ≈ Telfer setting Paterson South 864 2 gr / 7 apps Au, Cu ~45km SW of Paterson; on trend to Nifty & Maroochydore Cu Mt Cecelia ~135 1 granted Au ~130km NW of Paterson; ~70km W of Rio’s Winu Panorama ~105 3 granted Au, Ni East Pilbara greenstone; orogenic Au + Ni sulphide TOTAL ~1,780 11 gr / 7 apps Multi Cornerstone: Greatland (~11%) 5.1 Paterson — The Flagship If you are buying ARE, Paterson is the centre of gravity. Five granted licences covering 676 km² — Black Hills, Black Hills North, Paterson Range East, Budjidowns and Basel — sitting directly on trend with Greatland’s Havieron and Telfer assets (Figure 2). The ground is being acquired from Greatland and from a Greatland–Rio Tinto joint venture. Figure 2: The flagship Paterson Project combines favourable geology, walk-up drill targets, Telfer and Havieron analogues and untested structural positions (Source: The company Website) The most important point is that there are already gold hits on this ground, not just next to it. Historical drilling at Black Hills returned intercepts including 13m @ 2.01 g/t Au from 67m and 12m @ 1.38 g/t Au from 32m, plus deeper gold-copper anomalism (3.5m @ 1.88 g/t Au from 226.5m; 23m @ 706 ppm Cu). The Black Hills Dome has a geological setting compared to Telfer itself, and sits on the same regional structure as Antipa’s Minyari Dome. Budjidowns has thrown up scattered deeper hits (3m @ 0.52 g/t Au from 502m at Atlantis), and Paterson Range East and Basel both carry untested or partially tested targets under cover. The work plan is what you want to see: consolidate the large historical datasets, build 3D structural models, define drill-ready targets, run heritage surveys, then drill — RC and diamond across Black Hills, Paterson Range East, Basel and Budjidowns. Two of the tenements (Basel and Budjidowns) carry a 1.5% NSR royalty back to the Rio vendor. 5.2 Paterson South — Big, But Mostly Applications Paterson South is the largest project by area (864 km²) and the one to read most carefully. Only two of the nine tenements are granted — the rest are exploration licence applications, and Aventine has explicitly confirmed that no IPO proceeds will be spent on the applications. One application (ELA45/7247) overlaps a prior third-party application and, on grant, may be cut down to a much smaller area — possibly nothing at all. What’s there geologically is interesting: Throssell Range Group sediments including the Broadhurst Formation, which hosts the Nifty and Maroochydore sediment-hosted copper deposits. The granted ground (the Lamil JV tenement acquired from Rumble Resources and AIC Mines, plus an FMG–Carawine tenement) has had historical drilling that mostly failed to penetrate cover or test the targets Aventine is interested in. So the thesis is “under-tested, on a good address” — but the project is more option than near-term driver, and the application overhang is a real caveat. The Lamil tenement carries a 1% NSR to its vendors. 5.3 Mt Cecelia — The Best Single Intercept Mt Cecelia is a single granted licence (~135 km²) on the East Pilbara–Paterson boundary, acquired from West Wits Mining (ASX: WWI). It punches above its size because it has, arguably, the best historical gold result in the whole portfolio. After geophysics defined bedrock conductors, Rio Tinto Exploration drilled the project under a farm-in and hit: — WEWI0004: 24m @ 0.95 g/t Au within 82m @ 0.51 g/t Au from 128m — WEWI0001: 20m @ 0.93 g/t Au within 56m @ 0.55 g/t Au from 194m Those are wide, near-surface gold intercepts from a first-pass program — exactly the kind of result that justifies a follow-up drill campaign. The catch is the royalty and consideration load: Mt Cecelia carries two separate 1% NSR royalties (a combined 2%), plus the deferred milestone payment to West Wits — $1.0m in cash or 5m shares — that triggers on a 500,000-ounce JORC resource. That’s a lot of leakage stacked on one tenement, but it only bites if the project actually delivers. 5.4 Panorama — The Pilbara Optionality Panorama is three granted licences (~105 km²) of Archaean greenstone in the East Pilbara, prospective for orogenic (lode) gold and nickel sulphides (Figure 3). Soil and rock-chip sampling has flagged a 1.4km nickel anomaly (peak 0.3% Ni rock chip) and weak gold. No specific budget has been allocated to Panorama in the first two years — it’s held for early-stage reconnaissance and regional sampling. Treat it as free optionality, not a reason to subscribe. Figure 3: The Panorama Project is located west of Nullagine and south of Marble Bar, only 50km from Nullagine Gold Project. (Source: The company website) 06 / The Exploration Budget Here is the detail behind the ~$12m–$16.7m exploration line — and the single most encouraging feature is how much of it is drilling. Table 5 — Exploration Budget Category Yr 1 Min ($) Yr 2 Min ($) 2-Yr Min 2-Yr Max Access, heritage, tenure & licence 1,343,375 1,008,013 2,351,388 2,437,220 Soil sampling 84,425 — 84,425 82,500 Geophysics & consulting 432,000 120,000 552,000 669,500 Drilling & assays 2,093,026 3,083,318 5,176,344 8,496,492 Field operating costs 1,861,728 1,955,389 3,817,117 5,006,457 Grand total 5,814,554 6,166,720 11,981,274 16,692,169 Drilling and assays alone account for over half of the exploration spend — roughly $5.2m at minimum, scaling to ~$8.5m at maximum. That is the right shape for an exploration story whose entire thesis is “drill the targets the majors left behind.” Year 2 is materially more drilling-heavy than Year 1, which makes sense: Year 1 is about data consolidation, modelling, heritage clearances and first-pass drilling; Year 2 is where the follow-up campaigns land. One transparency note: the budget is presented by activity, not by project. Unlike some floats that tell you exactly how many dollars each asset gets, Aventine hasn’t published a per-project split. Management’s described priority work — drill-ready targets at Black Hills and the Mt Cecelia follow-up — tells you where the early focus sits, but the precise allocation is at the board’s discretion. 07 / The Board, the Management, the Story Behind Them A small-cap exploration IPO is a bet on people, and this is where Aventine is strongest. The standout thread is De Grey Mining and the Hemi discovery — both the Technical Director and the Exploration Manager come out of that camp, which is about as good a recent gold-discovery pedigree as exists in Western Australia. Table 6 — Board and Management Name Role Track record / signal Benjamin Dunn Managing Director 20+ yrs legal, equity & capital markets in resources. Senior roles at Citigroup, JP Morgan, CLSA. The capital-markets engine. Simon Andrew Non-Exec Chairman 20+ yrs financial markets across Australia & Asia. Exec Director of Mamba Exploration (ASX: M24), NED of Riversgold (ASX: RGL). Allan Kneeshaw Technical / Executive Director (Exploration) 30+ yrs geologist. Key role in De Grey’s 11Moz Hemi discovery. Prior: Anglo American, AngloGold Ashanti, Acacia, MIM. Competent Person. John Barbante Non-Exec Director 30+ yrs mining & business. Ex-CEO of White Tiger Resources, secured two major-company farm-ins. Government & industry network. Rowan Krasnoff Proposed Non-Exec Director The Greatland nominee, to be appointed after Admission. Rodney King Exploration Manager Ex-Principal Geologist at Northern Star (post De Grey acquisition), former De Grey Exploration Manager; a decade at Teck. Clarissa Chua Company Secretary Chartered Accountant; Co Sec for several ASX-listed resources companies (via Mining Corporate). At the prospectus date, the founding directors each held around 5% of a tiny share count; post-raise they sit near 1% each on shares, with the upside loaded into options (struck at or above the IPO price) and performance rights. That is a reasonable alignment structure — the team makes real money only if the stock re-rates well beyond $0.20. Table 7 — Director Interests on Admission (Maximum Subscription) Director Shares % (max) Options Perf. Rights Benjamin Dunn 1,625,001 1.0% 1,750,000 2,555,000 Simon Andrew 1,625,001 1.0% 1,600,000 1,462,500 Allan Kneeshaw 1,725,000 1.0% 1,750,000 2,555,000 John Barbante 1,625,000 1.0% 1,400,000 851,668 Rowan Krasnoff 250,000 0.1% — — Board total 6,850,002 ~4.1% 6,500,000 7,424,168 08 / The Greatland Factor The Greatland relationship is the most distinctive feature of this float, and like any cornerstone arrangement it is part validation, part transaction, part overhang. Table 8 — Substantial Holders on Admission Holder Shares (min / max) % (min / max) Greatland Resources (AIM; ASX: GGP) 15,865,154 / 18,642,932 11.4% / 11.1% West Wits Mining (ASX: WWI) 10,000,000 / 10,000,000 7.2% / 6.0% The Read Greatland Resources is not a passive name on the register. It owns and operates Telfer (3.2Moz Au) and Havieron (4.1Moz Au) — the two assets that define the modern Paterson Province. It is vending a large slice of Aventine’s tenure (the Paterson and Panorama ground, alongside a Rio Tinto JV interest), taking equity as part-consideration, becoming the company’s largest single shareholder at ~11%, and appointing a nominee director after listing. The bull reading is straightforward: the dominant operator in the district has looked at this ground, decided it is worth exploring but not worth holding inside its own portfolio, and chosen to back a focused vehicle to do the work — while keeping a seat at the table and equity upside. A Paterson Province specialist seeding a Paterson Province explorer is a meaningful signal. The bear reading is the mirror image, and it’s the same pattern you see whenever a major divests ground to an explorer: Greatland is offloading non-core, pre-resource tenure it chose to deprioritise behind Telfer and Havieron. The ground is early-stage, and Greatland’s continued involvement does not guarantee a discovery — it guarantees a knowledgeable, aligned shareholder. West Wits Mining, the Mt Cecelia vendor, sits alongside Greatland with ~6–7% and a deferred interest tied to a future resource. Both holdings are validation and, in time, potential supply. 09 / The Financials Aventine’s financials are exactly what you’d expect from a company incorporated in April 2025 and converted to a public company in April 2026. Table 9 — Financial Position Item Detail Incorporated 29 April 2025 (public company 3 April 2026) Operating revenue Nil (pre-revenue explorer) Existing cash (at Prospectus Date) $2,109,141 Total funds available post-raise (min / max) ~$17.1m / ~$22.1m Implied enterprise value at listing (approx.) ~$11m (market cap less available cash) Dividend policy None expected near term Two things to note. First, unlike many pre-IPO explorers that arrive on the eve of listing running on fumes, Aventine comes to market with ~$2.1m of cash already in hand — modest, but a more comfortable starting position than the bare-cupboard floats you often see. The prospectus presents the company as adequately funded for roughly two years on the back of the raise. Second, the implied enterprise value is the number that matters. Strip the IPO cash out of the ~$28–33.5m market cap and the market is effectively valuing the ground itself at around $11m. Most of the headline market cap, in other words, is the cash you’re putting in. That is normal for an exploration float, but it frames expectations correctly: you are paying ~$11m for a package of drill targets on a great address, not for a defined resource. 10 / The Market Context — A Great Address at a Great Time for Gold Most of the value case here is leverage to two things: the quality of the Paterson Province as a gold-copper address, and the gold price that funds the drilling. The Address The Paterson Orogen is one of the few Australian districts that has produced genuine tier-one discoveries in the modern era. Telfer has been a multi-decade gold producer; Havieron is one of the most significant gold-copper finds of the last decade; Rio Tinto’s Winu sits to the east; the Nifty copper mine and Antipa’s Minyari Dome round out the neighbourhood. Over in the East Pilbara, De Grey’s Hemi system — the discovery that several of Aventine’s team worked on — redefined what greenstone gold exploration could deliver and ended in a Northern Star takeover. The point is not that proximity equals a discovery (it never does), but that the ingredients, the structures and the host rocks that produced those deposits run through Aventine’s tenure. The gold backdrop The timing is, frankly, the easy part of the story. Gold has been trading around US$4,500 an ounce in mid-2026 — extraordinary levels by any historical measure, and a world away from the sub-US$2,000 environment of a few years ago. For a pre-revenue explorer, a strong gold price does two things: it makes capital easier to raise, and it makes even modest-grade intercepts look economically interesting. A 24-metre hit at ~1 g/t gold reads very differently at US$4,500/oz than it did at US$1,500/oz. That tailwind won’t last forever, and exploration risk is exploration risk regardless of the metal price — but it is hard to imagine a better moment to be funding a Paterson gold drill program. The copper angle (Paterson South’s Nifty-style targets, the copper anomalism at Black Hills) adds a second commodity string, into a market where copper’s structural supply story remains one of the most durable in resources. Samso Concluding Comments Aventine is a clean, well-constructed exploration float. There is no resource, no reserve, not even an exploration target — and the prospectus says so plainly. What you are backing is a drilling story on a genuinely good address, run by a team that has found gold before, and seeded by the district’s dominant operator. Most importantly, it is access to exploration tenure that has come from Greatland Gold - access that would not have happened without their participation. The things to like are real. The address is Tier-1. There are already gold hits on the ground — Black Hills and Mt Cecelia in particular give the team something concrete to chase rather than blue-sky nearology. The budget is lean and drilling-heavy, with over half of exploration dollars going into the drill bit. The Greatland cornerstone and board seat is the kind of validation money can’t manufacture. And gold near US$4,500/oz is the most supportive funding backdrop a gold explorer could ask for. The frictions are equally real, and they cluster around dilution and stage. Existing holders are diluted to ~19% on max — this is, in effect, a new company. Listing is conditional on completing five separate acquisition agreements, any of which could slip. A large slab of the Paterson South ground is still in application, with one tenement at risk of being granted over almost nothing. The royalty load across the portfolio is heavy — multiple 1% and 1.5% NSRs, plus the Mt Cecelia deferred consideration — which will matter if any project ever reaches production. And until the drill rig turns, every target in the portfolio is a hypothesis. The natural next conversation, exactly as with any quality explorer, is the first post-listing drill program — where it lands, what targets it tests, and whether Black Hills and Mt Cecelia convert historical hits into something that can grow toward a maiden resource. If they do, the re-rate path against a US$4,500 gold backdrop is obvious. If they don’t, this is a long, capital-hungry road like every other exploration story. The address is excellent and the team is credible; the deal is priced for exploration, and the drill bit will write the rest. Cover photo: Cetate open-pit gold mine, István Mihály. CC0, via Wikimedia Commons (https://commons.wikimedia.org/wiki/File:Cetate_open-pit_gold_mine.jpg) The Samso Way – Seek the Research Here at Samso, we pride ourselves on delivering content for investors that is independent and informed by over three decades of experience in the industry. Our content is well-researched and is only created if I see merit in discussing the company's story. Our mission is simple: cut through the noise and spotlight what matters—genuine stories, grounded insights, and real opportunity. Our content is well-researched and is only created if the team sees merit in discussing the company or concept. Investors can explore our three core platforms: Coffee with Samso Samso Insights Samso News There may be numerous paths to success in investing, but the common thread among successful individuals is that they remain committed to making informed decisions. Equip yourself with the right knowledge and tools, and you will be well on your way to achieving your financial goals. Most importantly, investors need to be absolutely diligent in understanding their own risk-reward tolerance and capabilities. Never bite off more than you can chew. As they say, Rome wasn’t built in a day, and the Great Wall stood because it took centuries to complete. The Samso Philosophy: Stay curious. Stay sharp. And remember—digging deeper always uncovers the real value. In Life, there is no such thing as a Free Lunch. Never bite off more than you can chew is my parting comment. Happy Investing, and the only four-letter word you need to know is DYOR. To support our independent nature of our work, please head over to our Support Page and give us a helping hand in any of the ways listed. This is a new initiative for the Samso Platform, and it was always the concept of Samso when we started this journey in 2018. Disclaimer The information or opinions provided herein do not constitute investment advice, an offer, or solicitation to subscribe for, purchase, or sell the investment product(s) mentioned herein. It does not take into consideration, nor have any regard to your specific investment objectives, financial situation, risk profile, tax position and particular, or unique needs and constraints. Read full Disclaimer. Share to Grow: Your Bonus Samso has just released an eBook: How to Add Value to your Share Portfolio |A lesson on geological models sought by mining companies that gives insight and an understanding of which portfolios are better - and potentially more lucrative – investments. Click here to download this eBook.| Download eBook If you find this article informative and useful, please help me share the information. I try to write about topics that are interesting and have the potential to be of investment value. It is not easy to find stories that fit those parameters. If you or your organisation sees the benefit of what Samso is trying to achieve and has a need to share your journey, please contact me at noel.ong@samso.com.au. About Samso Samso is a trusted platform that equips dedicated investors with up-to-date industry knowledge and insigh0ts from top CEOs and thought leaders. By staying informed on business advancements and market trends, investors can enhance their financial decisions through a combination of expert guidance and their own research. Samso News | www.samso.com.au | An Investor Lens on ASX-Listed Companies

An Australian drone-target maker scaling into a global counter-drone wave — already in the field with allied militaries. 01 / The 60-Second Pitch Boresight Ltd (proposed ASX: BST) is a Canberra-based defence-tech company listing on the ASX on 10 June 2026, having raised $8m at $0.20 per share. What makes it stand apart from the usual micro-cap float is simple: it already sells things to real customers. This is not a concept or a drill target — it’s an operating manufacturer with revenue, a product range, and a foothold inside allied militaries. The product is deceptively mundane and rather clever. Modern warfare has been reshaped by cheap drones — for surveillance, and increasingly as weapons. Militaries now have to buy and operate counter-drone systems, and then train against the threat. Training against real, expensive ISR drones is wasteful; you don’t want to shoot down a $50,000 aircraft to test a jammer. Boresight’s answer is a range of low-cost “attritable” target drones — built to be cheap enough to be shot out of the sky repeatedly, but capable enough to mimic the threats soldiers will actually face. More targets, less money, more repetitions, better-trained operators. The flagship is the BQ400 quadcopter target, backed by mission-planning software and a “swarming” ground-control system that lets one operator fly many targets at once — exactly the kind of complex, many-against-one scenario that C-sUAS crews need to rehearse. A fixed-wing target (the BF150) and a general-purpose quadcopter (ASCA) are in the pipeline. The company said its military customers include 11 Western armed forces, listing the Australian Defence Force, British Army, Canadian Armed Forces, New Zealand Defence Force, the United States Army, Marine Corps, Navy and Air Force, and the Dutch, Italian and Finnish armies. The pitch to investors is a scaling story: take a proven, in-demand product and a thin manufacturing base, add public-market capital, and build capacity — expand the Fyshwick (Canberra) operation and stand up a US plant in Alabama to chase the world’s largest defence market. The tailwind is real and politically durable: Australia’s updated Integrated Investment Program commits an estimated $12–15bn to drone and autonomous systems through 2036, with counter-drone explicitly in scope. What you’re buying, then, is early-stage exposure to a genuine business in a hot sector — with all the execution and concentration risk that “early-stage” implies. 02 / Boresight (ASX: BST) IPO Snapshot Table 1 — IPO Snapshot Item Detail Company Boresight Ltd (ACN 642 501 228) Proposed ASX Code BST Offer Price $0.20 per share Raise $8.0m (40,000,000 shares, before costs) Indicative Market Cap ~$41.8m (at offer price) Pro-forma Cash on Listing ~$8.94m FY25 Revenue $4.36m (FY24: $2.77m) FY25 Result Net loss of ~$0.60m Lead Manager CPS Capital Corporate Adviser ARQ Capital Share Registry Xcend Lodgement / Offer close / Quotation 23 Apr / ~19 May / 10 June 2026 Sector Defence technology — counter-drone (C-sUAS) training systems HQ / Operations Fyshwick, ACT (Canberra); US facility in Madison/Huntsville, Alabama 03 / Capital Structure & Dilution Here the contrast with a typical explorer float is stark — and in Boresight’s favour. The $41.8m indicative market cap at a $0.20 offer price implies roughly 209 million shares on issue after listing. The 40 million new shares sold in the IPO therefore represent only about 19% of the company. Existing holders aren’t being washed out; they’re bringing the public in for a minority slice and keeping the lion’s share. Table 2 — Capital Structure Security Shares % of company Existing shares (pre-IPO) 169,022,782 81% New IPO shares 40,000,000 19% Total on listing 209,022,782 100% That low dilution cuts both ways. It signals founders and early backers believe in the upside and want to keep it — good alignment. But combined with a heavily insider-held register (see Section 7), it also means the free float is small, which tends to mean thin liquidity and sharper price swings once trading begins. There’s also a meaningful options stack — the directors collectively hold well over 15 million options — which will sit over the stock as future dilution if exercised. Expect a chunk of insider stock to be escrowed for the usual 12–24 months; the prospectus has the detail. 04 / Use of Funds Boresight IPO has been refreshingly plain about what the $8m is for: making more drones, faster, in more places. The proceeds are earmarked to expand the engineering and production teams, ramp up production in the United States, increase additive manufacturing (3D printing) capacity, and further vertically integrate the operation — alongside expanding the existing Canberra facility and leasing the larger Alabama site. Table 3 — Use of Funds (as described) Use Purpose Manufacturing capacity Expand Fyshwick (Canberra); increase additive-manufacturing capacity US expansion Stand up / lease a larger facility in Madison–Huntsville, Alabama People Grow engineering and production teams Vertical integration Bring more of the supply chain in-house Working capital General corporate purposes The prospectus does not publish a precise dollar split across these line items; the above reflects the categories management has disclosed. The logic is sound: the binding constraint on a business like this isn’t demand, it’s the ability to build units at volume and at cost. The US footprint is the strategically interesting bit — Alabama (Huntsville is a major US defence-and-space hub) puts Boresight next to the world’s biggest defence buyer. It’s also where execution risk is concentrated and where export control and dual-use compliance become more complicated. 05 / The Business — What Boresight Actually Sells This is the section an explorer can’t write, because there’s a real product line to describe. Table 4 — Product & Service Range Offering What it is Status BQ400 Quadcopter The flagship — a cost-effective quadcopter target for C-sUAS training and testing (Figure 1) In production / in field Swarming GCS Ground-control + flight-management software letting one operator fly many targets at once In use Mission Planning Software Pre-defined flight profiles and escalating scenarios for repeatable training In use ASCA GP-UAS Quadcopter A general-purpose quadcopter platform “Coming soon” BF150 A fixed-wing target to replicate faster, fixed-wing threats “Coming soon” The thesis hangs on the attritable idea: targets cheap enough to be destroyed routinely, but realistic enough to train against the evolving threat. That combination — low unit cost plus credible threat emulation plus software to orchestrate complex scenarios — is the moat Boresight is claiming, and it’s why it frames itself as the only ASX-listed pure-play in counter-drone training systems. Figure 1: Boresight's BQ400. The flagship — a cost-effective quadcopter target (Source: Company Website) The pipeline matters too. A fixed-wing target and a general-purpose platform would broaden the addressable training scenarios beyond quadcopters, but both are still “coming soon” — i.e. not yet revenue. For now, the BQ400 and the software around it are the engine. 06 / Traction & Financials The reason to take Boresight seriously is the numbers it already has on the board. Table 5 — Financial & Commercial Snapshot Metric Figure FY25 revenue $4.36m FY24 revenue $2.77m Revenue growth ~+57% year-on-year FY25 net result Loss of ~$0.60m Pro-forma cash on listing ~$8.94m Units sold (since 2020) 6,000+ target drones Customer base 15 militaries across 12 countries Incorporated 2020 (spun out of Criterion Solutions) Two things stand out. First, the revenue is real and growing fast — a ~57% jump in a year, off a base that’s already meaningful for a company of this size. Second, the loss is small (~$0.60m), which says this isn’t a cash-incinerating moonshot; it’s a near-breakeven business that needs capital to grow, not to survive. Pro-forma cash of ~$8.94m post-raise gives it a comfortable runway to fund the expansion. The honest caveat is the shape of defence revenue. Government and military orders are lumpy and slow — they arrive as episodic contracts, not smooth subscription curves, and procurement timelines can stretch. A 57% growth rate is impressive but won’t necessarily repeat in a straight line; one delayed program can swing a half-year. Investors should watch the cadence of new orders post-listing more than any single revenue figure. 07 / The Board & Ownership A company this early is a bet on the people and the relationships behind it — and Boresight’s register is tightly held by its founders and the company it was born from. Table 6 — Board & Key Holders Name Role Background/holding (at prospectus) Justin Olde Managing Director & CEO Joined 2022; previously 4+ years as an executive at Electro Optic Systems (ASX: EOS). Held no shares but 8m+ options. Michael Sinkowitsch Co-founder, Executive Director Former Australian Army officer. Held ~23.5%; 3.2m options. Dr Andrew Windsor Non-Executive Chair (UK-based) Held ~21.7%; 3.2m options. Blake Burton Non-Executive Director (Perth) Held ~0.36%; 1.2m options. Criterion Solutions Substantial holder (related party) Canberra defence & intelligence company that Boresight was spun out of in 2020; held ~25%. Sinkowitsch and Windsor are directors/shareholders. The strength is domain pedigree: a co-founder with Army experience, a CEO out of a listed defence company (EOS), and a parent (Criterion) embedded in the Canberra defence-and-intelligence ecosystem. These are people who know the customer and the procurement world. The flip side is concentration and related-party exposure. Founders plus Criterion control a large majority of the register, and Criterion is both a major shareholder and the entity Boresight emerged from — a related-party relationship that warrants reading the prospectus’s disclosures on any ongoing arrangements. It also reinforces the thin-free-float point: with so much stock held by insiders (and likely escrowed), the freely traded portion is small. 08 / The Market — Why Now The macro case is the easiest part of the story, and it’s genuinely strong. Drones have changed warfare and public safety permanently. Cheap uncrewed systems are now used by sophisticated and unsophisticated adversaries alike, for surveillance and as weapons. That has forced militaries worldwide to acquire counter-drone systems — and, crucially, to train against the threat continuously. Counter-drone training is a structural, recurring need, not a one-off purchase, and that’s the slice Boresight occupies. The funding backdrop is concrete. Australia’s updated Integrated Investment Program earmarks an estimated $12–15bn for drone and autonomous systems through 2036, with counter-drone capability explicitly named. Globally, the lessons of recent drone-heavy conflicts have pushed C-sUAS up every allied military’s priority list. And Boresight’s planned US footprint in Alabama points it at the largest defence market in the world. A sober note: defence spending is politically driven and program timelines are long, so “the budget exists” doesn’t automatically convert to “Boresight wins the order.” But the direction of travel — more drones, therefore more counter-drone systems, therefore more training against targets — is about as durable a thematic as you’ll find in defence right now. 09 / The Risks / Points of Friction ▸ Lumpy, slow revenue. Defence procurement is episodic and politically timed; growth won’t be linear, and a delayed program can dent a reporting period. ▸ Still loss-making. Small, but the path to sustained profitability at scale is unproven. ▸ Concentration & related party. Insider- and Criterion-dominated register; small free float means thin liquidity and volatility, plus related-party considerations to read closely. ▸ Execution risk on US expansion. Standing up the Alabama facility, hiring, and scaling additive manufacturing are real operational challenges — and bring US export-control / dual-use compliance. ▸ Competition. The target-drone and C-sUAS space is busy; Boresight’s “attritable + realistic + swarming” niche is a claim that competitors will contest. ▸ Pipeline not yet revenue. The BF150 and ASCA are “coming soon”; today’s revenue leans on the BQ400 and its software. ▸ Valuation. ~$41.8m for ~$4m of revenue is a growth multiple — it prices in the scaling, so execution has to deliver. Samso Concluding Comments Boresight is the most “real” company in this run of listings — this is also the kind of Samso coverage that makes us different from the rest of the reviews online. This is something that makes money and revenue is the key to the discussion. There’s no resource to define, no drill program to wait for: there’s a product in the field, paying customers across a dozen countries, and revenue growing at a clip. For a $42m ASX float, that combination is genuinely uncommon. The things to like are tangible. The product solves a real and growing problem cheaply. The revenue is up ~57% and the loss is small, so the capital is for growth rather than survival. The sector tailwind — counter-drone training as a structural, recurring defence need — is strong and politically durable, and the US expansion points the company at the biggest market going. Founders are keeping most of the equity, which aligns them with new shareholders. The frictions are just as real, and they cluster around concentration and conversion. The register is tightly held by insiders and a related party, so liquidity will be thin and the free float small. Revenue is lumpy and procurement is slow, so patience is required. US expansion is where the money goes and where the execution risk lives. And the price already assumes the scaling works — at ~$41.8m on ~$4m of revenue, you’re paying for the next few years, not the last one. The natural thing to watch, as with any growth-stage listing, is the order flow after listing — new contracts, the ramp of the US facility, and whether the “coming soon” pipeline (BF150, ASCA) converts into revenue. If the orders keep landing and the Alabama plant comes online on plan, the thematic does the rest. If procurement stalls or the scale-up slips, this is a small, thinly traded defence stock priced for a growth that hasn’t yet been proven at volume. Real business, real tailwind, real execution risk — and the contracts will tell the story. Cover photo: US Navy 060621-M-8484J-009 A Pioneer Unmanned Aerial Vehicle (UAV) assigned to Marine Unmanned Aerial Vehicle Squadron Two (VMU-2) is ready to depart Al Taqaddum, U.S. Marine Corps photo by Sgt. Jennifer L. Jones. Public domain, via Wikimedia Commons (https://commons.wikimedia.org/wiki/File:US_Navy_060621-M-8484J-009_A_Pioneer_Unmanned_Aerial_Vehicle_(UAV)_assigned_to_Marine_Unmanned_Aerial_Vehicle_Squadron_Two_(VMU-2)_is_ready_to_depart_Al_Taqaddum.jpg) The Samso Way – Seek the Research Here at Samso, we pride ourselves on delivering content for investors that is independent and informed by over three decades of experience in the industry. Our content is well-researched and is only created if I see merit in discussing the company's story. Our mission is simple: cut through the noise and spotlight what matters—genuine stories, grounded insights, and real opportunity. Our content is well-researched and is only created if the team sees merit in discussing the company or concept. Investors can explore our three core platforms: Coffee with Samso Samso Insights Samso News There may be numerous paths to success in investing, but the common thread among successful individuals is that they remain committed to making informed decisions. Equip yourself with the right knowledge and tools, and you will be well on your way to achieving your financial goals. Most importantly, investors need to be absolutely diligent in understanding their own risk-reward tolerance and capabilities. Never bite off more than you can chew. As they say, Rome wasn’t built in a day, and the Great Wall stood because it took centuries to complete. The Samso Philosophy: Stay curious. Stay sharp. And remember—digging deeper always uncovers the real value. In Life, there is no such thing as a Free Lunch. Never bite off more than you can chew is my parting comment. Happy Investing, and the only four-letter word you need to know is DYOR. To support our independent nature of our work, please head over to our Support Page and give us a helping hand in any of the ways listed. This is a new initiative for the Samso Platform, and it was always the concept of Samso when we started this journey in 2018. Disclaimer The information or opinions provided herein do not constitute investment advice, an offer, or solicitation to subscribe for, purchase, or sell the investment product(s) mentioned herein. It does not take into consideration, nor have any regard to your specific investment objectives, financial situation, risk profile, tax position and particular, or unique needs and constraints. Read full Disclaimer. Share to Grow: Your Bonus Samso has just released an eBook: How to Add Value to your Share Portfolio |A lesson on geological models sought by mining companies that gives insight and an understanding of which portfolios are better - and potentially more lucrative – investments. Click here to download this eBook.| Download eBook If you find this article informative and useful, please help me share the information. I try to write about topics that are interesting and have the potential to be of investment value. It is not easy to find stories that fit those parameters. If you or your organisation sees the benefit of what Samso is trying to achieve and has a need to share your journey, please contact me at noel.ong@samso.com.au. About Samso Samso is a trusted platform that equips dedicated investors with up-to-date industry knowledge and insigh0ts from top CEOs and thought leaders. By staying informed on business advancements and market trends, investors can enhance their financial decisions through a combination of expert guidance and their own research. Samso News | www.samso.com.au | An Investor Lens on ASX-Listed Companies

The company recasts the board and executive team with Sierra Rutile Veterans as the Minta Rutile & Monazite Project Pushes Toward a Maiden Mineral Resource Estimate Lion Rock Minerals Limited (ASX: LRM), formerly Peak Minerals, announced one of the more striking leadership reshuffles seen on the small-cap mineral sands board this year, installing two former Sierra Rutile executives — Theuns de Bruyn as Chief Executive Officer and Grant Scott as Chief Operating Officer. In addition, the board will be bolstered by Duncan Craib, the former Boss Energy MD, alongside David Brophy, a West African Licensing Expert, as Non-executive directors. The reshuffle is bookended by a $2 million cornerstone placement at $0.02 per share and a clear roadmap toward a maiden Mineral Resource Estimate at the Minta Rutile & Monazite Project in the second half of 2026.- supported by the incoming directors and executives. Lion Rock has been somewhat quiet in the last 12 months, and while they have in the past presented a compelling story for their project in Cameroon, the company has been very shy in their communication with the market. The recent ASX release seems to be announcing a clear message to the market that Lion Rock is back to play again. A SIERRA RUTILE REUNION, TRANSPLANTED TO CAMEROON The central narrative has not changed since Samso started covering Lion Rock Minerals and this is a deliberate recreation of a Sierra Rutile-style operating model at Minta. Mr de Bruyn and Mr Scott previously worked together at Sierra Rutile Holdings Limited — then the world’s largest natural rutile producer — and between them carry more than 60 years of experience across project development, feasibility, financing, in-country operations and product marketing in the African mineral sector. Figure 1: The new Board and Executive team at Lion Rock Minerals (Source: Samso, compiled from LRM ASX announcement, 21 May 2026). Mr de Bruyn’s background spans host-government engagement, communities, and international development finance institutions. Mr Scott’s technical remit at Sierra Rutile covered surveying, mine planning, resource and reserve estimation, geotechnical and geological drilling, rehabilitation, water and tailings management, reconciliations and feasibility studies. Both CEO and COO will be based in-country in Cameroon for the duration of the Minta Project’s development phase. That pairing is now bolted onto a board led by Duncan Craib, whose recent career arc took Boss Energy Limited (ASX: BOE) from a microcap Australian holding company to an international ASX 200 uranium producer, and who served as Finance Director at Swakop Uranium during its US$2.2 billion corporate takeover and US$2.5 billion development of the Husab mine in Namibia, becoming one of the world’s largest uranium producers. Mr Craib also chaired the Uranium Forum of the Minerals Council of Australia from September 2021 to August 2025, and has successively overseen two international uranium projects through their growth phases into producing assets with worldwide sales distribution. He visited Cameroon twice before accepting the chairmanship — first as a guest to inspect the projects, and again with Mr de Bruyn and Mr Scott to undertake due diligence on the ground, including meetings with senior representatives of the Cameroonian Minister of Mines, Industry and Technological Development. David Brophy joins the Board as Non-Executive Director and is, in his own way, as deliberate a piece of the new structure as the Sierra Rutile pairing. His 20+ years of commercial experience spans West and Central African commodity supply chains, risk management, derivative hedging, and client-focused distribution across global markets. His career has run through senior roles at ECOM, Noble Group and Glencore Grain BV, where he contributed to the establishment of global cotton trading desks in Singapore and Rotterdam. Since returning to Australia, Mr Brophy has opened and developed West and Central African procurement operations across Mali, Burkina Faso, Benin, Ivory Coast and Cameroon — the latter being the jurisdiction in which Minta sits. He operates a resources consulting business assisting mining companies with licensing matters in West Africa, alongside a cotton procurement and logistics business focused in the region, and is the founder of Mansa Carbon, a project developer active in the Voluntary Carbon Market in West Africa. From a project assessment perspective, the Brophy appointment fills a specific gap. Where Mr Craib brings uranium-development and capital-markets pedigree, and Messrs de Bruyn and Scott bring the rutile-mining technical and operating pedigree, Mr Brophy brings the commercial, supply-chain and licensing relationships across West and Central African jurisdictions — specifically including Cameroon — that an asset like Minta will require as it progresses from exploration into permitting, offtake structuring and eventual product marketing. That is a useful complement to a board whose other appointees are concentrated on the technical and corporate dimensions of the story. Outgoing directors Robert Boston and Phillip Gallagher retire from the Board with immediate effect. The Board acknowledged Mr Gallagher’s particular contribution in identifying and securing the province-scale rutile and monazite opportunity at Minta — the initiative that established the foundation on which the Company’s next phase is being built — and thanked Mr Boston for his contribution to the Company’s formation and governance over the last nine years. The structural decision that matters here is the relocation of both the CEO and COO to Cameroon for the development phase. The 14 March 2026 management update flagged that the prior Board had concluded a greater in-country presence was needed to advance exploration and development at Minta. The new appointments deliver that change in literal form — not just senior oversight, but day-to-day technical decision making, government engagement and operational control located alongside the asset. The team has started preliminary work while visiting the project on that second visit. The incoming team inspected active drilling areas, reviewed the in-country laboratory facility being commissioned to support field sampling, and visited Minta Est, where field panning and visual inspection had indicated the presence of heavy minerals including rutile, ilmenite, zircon and monazite. The team also met with senior representatives of the Cameroonian Minister of Mines, Industry and Technological Development. Management Commentary — The Strategic Context “Our Minta Project is attracting world-wide attention for its significant and growing resource base of rutile and monazite.” — Duncan Craib, Incoming Non-Executive Chair, Lion Rock Minerals Limited (21 May 2026) “Lion Rock’s Minta Project has the hallmarks of a globally significant rutile and monazite producer.” — Theuns de Bruyn, Incoming Chief Executive Officer, Lion Rock Minerals Limited (21 May 2026) THE MINTA RUTILE & MONAZITE PROJECT — THE ASSET THAT JUSTIFIES THE RESET The Minta Rutile & Monazite Project is Lion Rock’s flagship asset, comprising 18 granted exploration permits and three further permits under valid application across approximately 8,800 km² of prospective ground in central Cameroon, with the 21 May 2026 announcement describing approximately 5,000 km² of prospective ground in the higher-priority belt within that footprint. Lion Rock holds an 80% group interest in each of the Minta tenements. Mineralisation styles include the residual concentration of valuable heavy minerals — monazite, rutile and zircon — in deflated soils across elevated terrain, alongside the concentration of valuable heavy minerals and gold in large-scale alluvial basins. In-situ vein-hosted gold has been artisanally mined and mapped and sampled by the Company and remains under consideration for focused exploration. A short Samso primer on the heavy minerals at Minta Rutile is the principal natural form of titanium dioxide (TiO₂) and the highest-grade titanium feedstock used in the production of titanium pigment, titanium metal and welding fluxes. Natural rutile typically carries a TiO₂ content above 90%, which is why integrated producers like Tronox value direct access to rutile feedstock streams. Monazite is a rare-earth-bearing phosphate mineral, typically rich in light rare earth elements (cerium, lanthanum, neodymium, praseodymium) but also carrying a measurable component of heavy rare earths via the coexistence of xenotime. Monazite-rich heavy mineral concentrates are an established feedstock for rare-earth processing routes. Valuable Heavy Mineral (VHM) is the industry shorthand for the saleable portion of a heavy mineral assemblage — in the Minta context, that means rutile, zircon and monazite. The "dual residual–alluvial" strategy that Lion Rock has flagged is essentially a recognition that both the in-situ residual soils and the downstream alluvial basins can host economic VHM concentrations. Heavy-liquid separation (HLS) is a laboratory technique that uses a dense fluid to physically separate heavy minerals (rutile, monazite, zircon, ilmenite) from lighter gangue minerals. Bringing HLS in-country at the Yaoundé laboratory materially compresses the loop between drilling, mineralogical interpretation and product qualification work. Geologically, the central Cameroon rutile system is interpreted to have crystallised from the scavenging of titanium-bearing units from sediments subjected to high temperatures and pressures during regional metamorphism. At least two generations of granite intrusion have introduced fluids and heat that remobilised gold and introduced monazite and zircon at Minta Est. The coarse rutile crystals identified across broad areas indicate the late emplacement of a broad-scale pegmatite vein system — and large angular rutile nuggets identified in recent and historical sampling programs have the potential to materially boost total Valuable Heavy Mineral grade in residual and alluvial prospects. Zones of very high-grade zircon mineralisation have also been identified at Minta Est, alongside alluvial and hard-rock gold occurrences across the northeastern tenement area coincident with a geophysical anomaly associated with granitic intrusions. Figure 2: The Company’s flagship Minta Rutile and Monazite Project in Cameroon (Source: LRM ASX announcement, 21 May 2026). The infrastructure setting matters for any West or Central African development story — and the incoming team was specific about its endorsement of what it found. The 21 May 2026 announcement described the regional infrastructure setting as featuring maintained road access, nearby rail connections via the Nanga Eboko rail line, power supply and internet access, all of which contribute to the project’s development pathway. Mr de Bruyn extended that view to the country itself. “The country is more organised, connected and capable of supporting mining development than many outside observers might expect.” — Theuns de Bruyn, on the Cameroon operating environment (21 May 2026) Strategic Partnership with Tronox Sitting behind the operational restructure is an existing relationship with Tronox Holdings plc (NYSE: TROX), an integrated producer of titanium dioxide with US-government endorsement of its proposed rare earth refinery. Tronox has already secured a 5% interest in Lion Rock and has identified Minta as a potential feedstock source for its rare earth strategy. The 21 May 2026 announcement frames the partnership as focused on accelerating the Minta Project and aligning with Tronox’s strategy to secure end-to-end rare-earth supply chains. From a project assessment perspective, this is the strategic feature of the Minta story that most distinguishes it from the rest of the ASX rutile small-cap field. A NYSE-listed integrated TiO₂ producer with declared rare-earth ambitions does not put 5% on a register without an operational thesis. Whether that thesis converts into a formal offtake or a deeper equity move over the next twelve months is one of the structural questions overhanging the Minta story. OPERATIONAL TEMPO During the March 2026 quarter, Lion Rock completed 681 holes for 3,310 metres of infill drilling at high-value rutile targets across the wider Minta Project. A focused 299-hole, 1,281-metre infill program was also completed at Minta Est over the ~250 km² monazite-enriched granite that the Company has identified as the source of monazite, xenotime and zircon coincident with high-grade rutile zones. Figure 3: Ongoing infill drilling activities at Minta Est (Source: LRM March 2026 Quarterly Activities Report, 28 April 2026). On the assay side, further high-grade rutile results received during March 2026 confirmed in-situ rutile grades of up to 2.6% from residual targets, with multiple locations returning at or above 1.0% rutile and a broader population at or above 0.5%. Standout intercepts included up to 4 metres at 1.8% rutile from residual drilling. Results spanned both residual and alluvial settings, supporting the Company’s dual residual–alluvial development strategy across a >5,000 km² footprint, with key alluvial corridors — including the Yong river basin — flagged as priority follow-up targets. Figure 4: Minta Est infill drilling coverage on thorium radiometrics as at January 2026, with monazite assemblage results from completed drill holes (Source: LRM March 2026 Quarterly Activities Report). The spatial context of the Minta Est infill is what makes the program material. The Company’s January 2026 drilling coverage map (Figure 4) shows the infill drill pattern superimposed over thorium radiometrics, with individual completed drill holes labelled with their monazite assemblage percentages — ranging from the 1.9% recorded at ME22S07 up to a standout 73.7% monazite assemblage at ME22S04. The high-monazite-assemblage holes (ME22S04, ME22S03, RE1028 and RE0014) cluster within the granite footprint that the Company has identified as the local source of monazite, xenotime and zircon, providing the geological basis for the dual residual–alluvial development strategy. Table 1: Selected heavy mineral results from Minta and Minta Est (Source: LRM March 2026 Quarterly Activities Report) Drill hole / Location Mineral / Metric Value Context ME22S04 Monazite assemblage 73.7% Minta Est, monazite-enriched granite domain ME22S03 Monazite assemblage 40.0% Minta Est, central granite domain RE1028 Monazite assemblage 37.2% Minta Est, residual target RE0014 Monazite assemblage 35.6% Minta Est, residual target ME22S02 Monazite assemblage 25.1% Minta Est, granite domain ME22S01 Monazite assemblage 18.0% Minta Est, southern granite domain ME22S05 Monazite assemblage 14.3% Minta Est, eastern granite domain RE0018 Monazite assemblage 12.5% Minta Est, residual target ME22S07 Monazite assemblage 1.9% Minta Est, granite margin Residual targets — multiple Peak in-situ rutile Up to 2.6% Multiple locations ≥ 1.0% rutile Residual drilling — standout In-situ rutile intercept 4 m @ 1.8% Wider Minta high-value targets Monazite assemblage values represent the percentage of monazite within the heavy mineral assemblage at the labelled drill holes as published in the Company’s January 2026 coverage map (Figure 4). In-situ rutile results refer to grades published in the Company’s 18 March 2026 release. Assays from the broader 299-hole Minta Est program and the 681-hole rutile program were pending at the time of the March 2026 Quarterly Activities Report. From a project assessment perspective, the most important spatial outcome of the March quarter is the demonstration that the monazite-enriched granite at Minta Est is locally sourcing the heavy mineral assemblage. The presence of coarse, angular monazite crystals in residual soils — with limited transport — reinforces the case for both residual concentration in deflated high-ground domains and alluvial accumulation in downstream basins. That dual style is what underpins the maiden Mineral Resource Estimate workflow now in train. Laboratory and Mineralogical Workstreams The Yaoundé in-country laboratory — currently being commissioned with heavy-liquid separation capability — is one of the more underrated structural decisions in the recent operating playbook. It is designed to support ongoing assay prioritisation, mineral assemblage studies and future process flowsheet design while reducing reliance on offshore sample preparation and allowing faster iteration between drilling and mineralogical interpretation. For an exploration program built around throughput and a maiden MRE target in H2 2026, that turnaround time matters. Path to the Maiden Mineral Resource Estimate Pending assays from the 299-hole Minta Est program and the broader 681-hole rutile program are expected to feed into geological modelling and resource estimation workflows during H1 2026, with the maiden Mineral Resource Estimate targeted for H2 2026. The Company has framed the next twelve months as the value-defining exploration phase — a phrase that captures the fact that the market does not yet have a JORC-compliant resource against which to value the district-scale story. Figure 5: Minta at a glance — March 2026 quarter operating metrics, Placement structure, and the 6–12 month milestone roadmap (Source: Samso, compiled from LRM ASX announcement of 21 May 2026 and LRM March 2026 Quarterly Activities Report). THE $2 MILLION CORNERSTONE PLACEMENT — FUNDING THE NEXT PHASE Lion Rock has received firm commitments to raise $2 million through a placement of 100 million new shares at $0.02 per share, conducted under the Company’s existing placement capacity pursuant to ASX Listing Rule 7.1A. New shares rank equally with existing shares on issue. The Placement was cornerstoned by the incoming directors and executives, and supported by a limited number of known and respected industry professionals. Following settlement, the Company will have a pro forma cash position of approximately $8 million (comprising the $2 million capital raise plus existing cash reserves of $6 million as at 15 May 2026). Placement funds are earmarked for ongoing exploration at the Minta Project, business development activities and general working capital. The pro forma quarters of funding at the March 2026 burn rate (~A$1.26m on exploration during the quarter) would extend the Company’s runway well beyond the targeted maiden MRE date. Incoming Team Incentive Package Under the terms of the agreements with the new leadership team, the Company has agreed to issue a total of 270 million unquoted options structured as follows: Table 2 : Incoming Team Incentive Package Tranche Quantity Exercise Price Expiry Incoming team — Tranche A 135,000,000 options $0.03 3 years from issue Incoming team — Tranche B 135,000,000 options $0.04 3 years from issue CEO incentive (de Bruyn) — Tranche A 30,000,000 options $0.03 21 May 2029 CEO incentive (de Bruyn) — Tranche B 30,000,000 options $0.04 21 May 2029 Source: LRM ASX announcement, 21 May 2026. The CEO incentive options form part of the Company’s Employee Securities Incentive Plan. Mr de Bruyn’s contract carries an annual salary of A$450,000 with a three-month probationary period (one-month termination notice during that period) and a three-month termination clause thereafter. Table 3: Capital Structure (at 31 March 2026 quarterly) Security Number on issue Ordinary fully paid shares 3,463,917,147 Ordinary fully paid shares — escrowed to 7 July 2026 100,000,000 Unquoted options @ $0.005 (exp. on/before 25 Sept 2027) 90,000,000 Unquoted options @ $0.0165 (exp. on/before 10 Feb 2028) 40,000,000 Unquoted options @ $0.020 (exp. on/before 10 Feb 2028) 40,000,000 Unquoted options @ $0.0165 (exp. on/before 9 May 2028) 46,600,000 Source: LRM March 2026 Quarterly Activities Report. The 21 May 2026 announcement adds a further 100 million new ordinary shares (Placement) and 270 million unquoted options to the incoming leadership team upon settlement and issue. OTHER PRIJECTS - KITONGO AND LOLO URANIUM PROJECTS — THE URANIUM OPTIONALITY Beyond Minta, Lion Rock retains the Kitongo and Lolo Uranium Projects in Cameroon. The Kitongo Uranium Project spans approximately 2,200 km² across five tenements in the northwest Adamoua Province; the Lolo Project covers approximately 240 km² across one tenement in the South region, approximately 70 km southwest of Yaoundé and 111 km from the Kribi deep-water port. All six tenements are currently pending grant. Figure 6: Location map of the Kitongo and Lolo Uranium Projects ( Source LRM Website) Prospectivity for uranium has been confirmed by significant historical exploration, initially in the 1970s and more recently between 2007 and 2011, when systematic exploration by Mega Uranium Ltd (TSX: MGA) — following its CAD150 million acquisition of Nu Energy Corporation in April 2007 — produced encouraging results from drilling programs undertaken in 2008 and 2010. During March 2009, Mega Uranium released results from an 11-hole diamond drilling program at Kitongo that indicated the presence of high-grade uranium mineralisation; in 2010, Mega Uranium completed a diamond drilling program at Lolo that tested a small portion of the 80 km-long prospective uraniferous belt and confirmed the presence of high-grade uranium mineralisation. During the March 2026 quarter, the Company continued working with uranium competent person Dr Marat Abzalov to design upcoming exploration programs while awaiting finalisation of the permitting process. The Company has explicitly noted that it has not been able to verify any of the historical drill intercepts and treats them as indicative only, pending confirmation drilling. From a project assessment perspective, the most interesting feature of the 21 May restructure is that Lion Rock now has — for the first time — a chairman with a direct uranium-development track record sitting above an asset that, on its face, looks like classic uranium-developer optionality. Mr Craib’s Boss Energy (ASX: BOE) and Swakop Uranium experience aligns directly with the kind of work program that Kitongo and Lolo will need once the permits are granted. That alignment is one of the quietly notable features of the May 21 restructure. PORTFOLIO HOUSEKEEPING — GREEN ROCKS DIVESTED The March 2026 quarterly also recorded the agreement to divest the Green Rocks Project tenements in Western Australia to Meekatharra Minerals East Pty Ltd (an unrelated party) by way of the sale of 100% of the shares in Greenrock Metals Pty Ltd and CU2 WA Pty Ltd for A$100,000 each (A$200,000 total). Completion occurred in April 2026. The Company determined the tenements to be non-core given their holding costs and minimal planned future work — a tidy-up that simplifies the corporate story ahead of the new team’s arrival. The Yendon Kaolin Project in Victoria (four licences in the Ballarat-Bendigo zone of the Lachlan Fold Belt) remains, with desktop review activities continuing during the quarter. NEAR-TERM MILESTONES TO WATCH Reading the 21 May 2026 announcement and the March 2026 Quarterly Activities Report side-by-side, the following near-term milestones have been disclosed across the two source releases: Table 4: Near-term milestones for Lion Rock Minerals Activity Timing Status / Source Settlement of $2m Placement (100m shares @ $0.02) Following 21 May 2026 announcement Firm commitments received; Placement under ASX Listing Rule 7.1A (21 May 2026 release) Issue of 270m unquoted options to incoming team On commencement 135m @ $0.03 and 135m @ $0.04; all 3-year expiry (21 May 2026 release) Cornerstone in-country relocation — CEO and COO Effective immediately Both based in Cameroon for the development phase (21 May 2026 release) Assays from 299-hole Minta Est infill program Q2 2026 1,281 m drilled over the monazite-enriched granite; assays pending (March 2026 quarterly) Assays from broader 681-hole, 3,310 m rutile program H1 2026 Assays feeding into geological modelling and resource estimation workflows (March 2026 quarterly) In-country Yaoundé laboratory commissioning H1 2026 Heavy-liquid separation capability being commissioned (March 2026 quarterly) Maiden Mineral Resource Estimate at Minta H2 2026 Headline catalyst signalled by the Company (March 2026 quarterly; 21 May 2026 release) Kitongo & Lolo Uranium Projects — permitting Pending All 6 tenements currently pending grant (March 2026 quarterly) Upcoming uranium exploration program design Underway Working with competent person Dr Marat Abzalov (March 2026 quarterly) ABOUT LION ROCK MINERALS LIMITED Lion Rock Minerals Limited (ASX: LRM), formerly Peak Minerals Limited, is an Australian-listed mineral sands and critical minerals explorer focused on the development of its flagship Minta Rutile & Monazite Project in central Cameroon. The Company holds an 80% group interest in 18 granted exploration permits and three further permits under valid application across approximately 8,800 km² of prospective ground at Minta, with approximately 5,000 km² described as the higher-priority prospective belt within that footprint. Beyond Minta, Lion Rock holds the Kitongo and Lolo Uranium Projects in Cameroon (six tenements totalling approximately 2,440 km², all currently pending grant) and the Yendon Kaolin Project in Victoria (four licences in the Ballarat-Bendigo zone of the Lachlan Fold Belt). The Green Rocks Project in Western Australia was divested in April 2026 as non-core. Tronox Holdings plc (NYSE: TROX), an integrated producer of titanium dioxide with US-government endorsement of its proposed rare earth refinery, holds a 5% interest in the Company and has identified Minta as a potential feedstock source for its rare earth strategy. The Company’s strategic objective is to deliver a maiden Mineral Resource Estimate at Minta in H2 2026 under the operational direction of a newly appointed Sierra Rutile-pedigree executive team, with both the CEO (Theuns de Bruyn) and COO (Grant Scott) based in-country in Cameroon for the duration of the Minta Project’s development phase. The Board is chaired by Duncan Craib, the former Managing Director and CEO of Boss Energy Limited (ASX: BOE), with David Brophy as Non-Executive Director bringing 20+ years of commercial experience across West and Central African commodity supply chains, licensing and procurement. As at the March 2026 quarterly, Lion Rock reported A$6.872 million in cash and cash equivalents, with a pro forma cash position of approximately A$8 million following settlement of the $2 million Placement announced on 21 May 2026. The Company’s capital structure at 31 March 2026 comprised 3,463,917,147 ordinary fully paid shares on issue (plus 100 million escrowed) and 216,600,000 unquoted options on issue. SAMSO CONCLUDING COMMENTS Samso has been covering the Lion Rock Minerals story for a long. We have also been a shareholder of the story and have since exited the majority of the holdings; however, this new management brings a different complexion to the story. One of the biggest issues with Lion Rock Minerals is the high market capitalisation. At the current share price, Lion Rock Minerals is valued at AUD $92M. This kinds of scare real punters way and I have to say that at that valuation, there are safer bets in the market. However, does that mean that this is not an investable story for the average punter on the ASX? This is a unique situation of having what looks like a classy project, and it does have very good backers, long-term and very influential investors. I think there will be a revaluation coming soon, and the new team will be looking to move on from this valuation; I have no doubt. The time for the sleeping giant to wake up may be imminent. DYOR as always, but look seriously. Cover photo: Rutile - Rutile, Nessa Eull. CC0, via Wikimedia Commons (https://commons.wikimedia.org/wiki/File:Rutile_(GeoDIL_number_-_1577).jpg)

Abstract Terra Metals (ASX: TM1) has re-rated roughly four-fold over the past twelve months, and the reason is not its defined resource. The company's only JORC estimate, the 148 Mt Dante Reefs deposit, is a bulk, low-grade titanium-vanadium system in a market that has shown little appetite for either metal at present. What moved the share price is Southwest, a high-grade platinum-group-metal, copper, nickel and cobalt sulphide discovery that sits entirely outside the Dante resource and, as yet, carries no resource of its own. This report sets out to answer a direct question. What is Terra Metals actually selling, and is the market right to price it where it does? Dante Reefs is a genuine bulk deposit, and its grades hold up well against producing vanadium-titanium-magnetite operations such as Panzhihua, but on stage it is the least advanced of its peer group. Skaergaard, Mount Peake and Mont Sorcier have all progressed further toward a study or feasibility. Southwest is the more interesting story. A discovery hole that returned 35 m at 2.90 g/t PGE3, with a peak of almost 53 g/t, sitting in the same geological family as Chalice's Julimar and BHP's Nebo-Babel. Both comparisons carry real weight, and both carry a caution. For every Julimar, the ASX has produced numerous discoveries that raised capital on an early result and never converted it into a defined resource. Metallurgy is the question that most conditions value here, for both assets. Dante's flowsheet recovers a titanium-ilmenite concentrate by magnetic separation, sidestepping the smelting hurdle that has stranded other vanadium-titanium projects, though the concentrate is not yet confirmed at a saleable grade and testwork remains at bench scale. Southwest's ultramafic host brings a different, well-documented problem, namely magnesium-silicate gangue, which has complicated processing at comparable nickel sulphide systems elsewhere. Neither question is resolved, and this report works through the published evidence on each rather than assuming either away. A broker note published in May already puts a fair value of A$0.84 on the stock, implying "205% upside" (East Coast Research, 6 May 2026), yet Terra has no scoping study, no PFS and no feasibility on either asset, only a resource and a discovery. I put no price target and no per-share valuation on the company here. With no economic study to build one from, a number would be invented rather than earned, and I would rather tell you plainly what is known and what is not. The prices that will ultimately decide the economics, platinum, titanium and vanadium among them, are also shaped as much by decisions in Beijing as by a free market, which this report sets out in full. Management carries genuine capital-markets and development experience, including a former Perseus Mining chief executive on the board, the kind of team a company that must fund a development pathway will need. The Samso Call is Watch and DYOR. The first deep diamond assays have now returned (21 July 2026) and held grade to 448 m, with two further releases since (20 and 26 August 2026) extending the corridor again. Roughly 32,000 m of assays remain pending, and the settling event is still the maiden Southwest resource targeted for late 2026, alongside Phase 2 metallurgical results on Dante. Until those results land, both the bull case and the bear case remain live, and nothing in the public record yet decides between them. Review the facts as they arrive, take a balanced view, and reassess as the picture sharpens. Contents 1 Why This Company Warrants a Full Workup 9 The Market: What Is TM1 Selling Into? 2 The Company & The Story 10 How Good Is Dante Reefs: And Is It Good Enough? 3 The Assets: The Numbers Themselves 11 The Metallurgy of Dante Reefs: Can These Ores Be Processed? 4 Comparable Deposits: How Dante Reefs Sits Globally 12 Comparable Deposits: Southwest (Ni-Cu-PGE Sulphide) 5 The Newsflow Since the Broker Report 13 The Metallurgy of Southwest: A More Conventional Route, With One Catch 6 The Broker's Number, and the Potential Upside 14 China and the Price of What TM1 Sells 7 What Would Change Our Mind 15 Samso Concluding Comments 8 Management 16 Disclaimer Terra Metals Limited - Snapshot Prices below are as at the date in the meta line above, the point this workup was compiled from. This is a living document. The share price and market cap will have moved, up or down, in the time since (check a live quote before acting on anything here, and see the note on data currency in the disclaimer). TABLE 1. TERRA METALS LIMITED, KEY METRICS Metric Value Share price A$0.235 (ASX, live at 27 Aug 2026; +2.17% on day, prices delayed ~20 min) Market cap A$237.10m (ASX, 27 Aug 2026) Cash A$75.1m at 30 June 2026 (June 2026 quarterly, Appendix 5B) Shares on issue 1,030.9 m (ASX, 27 Aug 2026) 52-week range A$0.080 to A$0.47 (ASX trailing 52-week window, 27 Aug 2026) 1 year ~+400% (13 Jul 2026 basis; not recomputed, current price is little changed) 2026 YTD ~+68% (13 Jul 2026 basis; not recomputed, current price is little changed) Since 13 Jul roughly flat (A$0.232 → A$0.235) despite three positive Southwest releases in that window, still ~50% off the A$0.47 peak Major broker cover None Substantial holders (Market Index, current). Golden Energy & Resources 14.90%, Tribeca 13.57%, Pine Energy 9.10%, M. Otway 8.69%. Share price context: a momentum stock that rose to A$0.47, fell ~31% in the month to 13 July 2026, and has since traded roughly flat through six weeks of positive Southwest newsflow, at A$0.235 as at 27 August 2026. (ASX, 27 August 2026.) 1 WHY THIS COMPANY WARRANTS A FULL WORKUP The starting point is the share price. Terra Metals is up approximately 400% over twelve months, and its market capitalisation stands at about A$237 million (ASX, 27 August 2026). A re-rating of that size demands an explanation, and the question this report sets out to answer is a direct one. What are investors actually pricing? It is worth being precise about what Terra owns today. Its only defined resource is the 148 Mt Dante Reefs deposit, whose principal contained metals are titanium and vanadium. Those are not, at present, strong markets (titanium feedstock demand is muted, and vanadium has been weak on soft steel demand), and a remote, bulk, low-grade titanium-vanadium project is not the kind of asset that typically drives a four-fold re-rating. On the evidence of the share price, the market is not paying for the titanium and vanadium. What changed is the Southwest discovery, a high-grade platinum-group-metal, copper, nickel and cobalt sulphide system that sits entirely outside the Dante resource. That is a different prize, exposed to a very different set of end-markets. This workup sets out to answer three questions. What is Terra ultimately selling? Is the Southwest prize substantive enough to justify the re-rating? And is the market right to price the company where it does? A broker note already assigns a fair value of A$0.84 and "205% upside" (East Coast Research, 6 May 2026), yet Terra has no scoping study, no PFS and no feasibility, only a resource and a discovery. We take up the underlying market, what investors are really buying, directly in Section 9. 2 THE COMPANY & THE STORY Plate 1. Diamond drilling at the Dante Project, West Musgrave. Source. Terra Metals, Tribeca Future Facing Symposium 2026 presentation, ASX release 25 August 2026. Terra Metals is a Western Australian critical-minerals explorer, 100% owner of the Dante Project in the West Musgrave region, hosted in the Jameson Layered Intrusion, part of the Giles Complex, a district-scale mafic-ultramafic system running ~80 km long and ~30 km wide under the company's tenure (TM1 ASX release, 1 July 2026). That is the same broad family as South Africa's Bushveld Complex, and Dante sits ~15 km north of BHP's Nebo-Babel, which the company notes was the largest discovery of this sulfide style since Voisey's Bay, before Julimar/Gonneville in 2020. Figure 1. Dante Project tenure over West Musgrave regional geology. The Dante Reefs Mineral Resource (148 Mt at 1.38% CuEq), the Southwest PGE-Cu-Ni discovery, and neighbours BHP's Nebo-Babel and Succoth. Source. Terra Metals, ASX release 25 August 2026. The project has two distinct parts. Dante Reefs is broad, near-surface and bulk-scale, but low-grade. Southwest is a high-grade magmatic sulfide system, and it is what re-rated the stock. Terra reported a maiden Dante Reefs resource in August 2025. It remains, in the company's own language, pre-scoping. That distinction underpins this entire report. DANTE REEFS: defined bulk Ti-V resource. 148 Mt @ 14.8% TiO₂, 0.54% V₂O₅, near-surface, pre-scoping. The established base. SOUTHWEST: high-grade sulphide discovery. No resource yet, sits outside the estimate, up to 52.97 g/t PGE3, open at depth. The source of the re-rating. Chart 1. Samso Research, from TM1 ASX releases (11 Aug 2025 MRE; 13 Jul 2026). 3 THE ASSETS: THE NUMBERS THEMSELVES Dante Reefs. The established base The August 2025 JORC estimate defines a near-surface polymetallic resource of 148 Mt at 14.8% TiO₂, 0.54% V₂O₅, 0.18% Cu and 0.33 g/t PGE3 (1.38% copper-equivalent), for roughly 22 Mt TiO₂, 800 kt V₂O₅, 270 kt copper and 1.6 Moz PGE3 of contained metal. It gives the company scale and commodity diversity. However, the material is low-grade and bulk, and the project is remote, so it would require very large tonnages, substantial capital and new infrastructure to become a mine. It is the established base of the project, not the driver of the upside. TABLE 2. DANTE REEFS MINERAL RESOURCE, AUGUST 2025 JORC ESTIMATE Southwest. The discovery Plate 2. Field crew and rig support at the Dante Project. Source. Terra Metals, Tribeca Future Facing Symposium 2026 presentation, ASX release 25 August 2026. Southwest is described by the company as a completely new style of magmatic sulfide mineralisation for the district. Not constrained to a single reef, spanning mafic to ultramafic rocks over far greater thickness than the Dante Reefs, with recent PGE7 assays showing iridium-group enrichment that points to a feeder-proximal system (TM1 ASX release, 1 July 2026). Crucially, it sits outside the 148 Mt Dante resource. Two translations for non-geologists. PGE3 is the combined grade of platinum, palladium and gold in grams per tonne. A gram per tonne is one part per million, so these are trace metals whose economics live on recovery and payability. "ended in mineralisation" means the drill stopped while still in the orebody. The discovery hole established the system. At SW5, SWT008 returned 35 m at 2.90 g/t PGE3 from 48 m, with a peak 1 m at 52.97 g/t PGE3. Through 2026, drilling has focused on establishing scale. The corridor between SW5 and SW6 was defined at that point over more than 950 m of strike, approximately 700 m of width and at least 750 m of down-dip extent from surface, within a package 150 m to more than 500 m thick and open in all directions. The 13 July 2026 results also confirmed the system reaches near-surface (SWRC064 returned 157 m of mineralisation from just 5 m downhole), which the company states supports the potential for a low strip-ratio starter pit (TM1 ASX releases, 18 June, 1 July & 13 July 2026). Drilling since has kept extending the system rather than closing it off. A 21 July 2026 release reported the first deep diamond assays, with SWRD051 returning 448 m at 0.87 g/t PGE3 and 0.15% nickel from 102 m (nested at 36.55 m at 2.00 g/t PGE3), the first real confirmation that the corridor holds grade at depth rather than only near surface. On 20 August 2026 the company reported a further 150 m of strike extension, taking the corridor to more than 1,100 m and describing it as still open. And on 26 August 2026 (the most recent release at the time of writing) further assays, including 392 m of combined mineralisation at SWRD057, confirmed the system continues to grow rather than close off. TM1 states it has now drilled more than 56,000 m at Southwest, with roughly 32,000 m of assays still pending as it works toward the maiden resource (TM1 ASX releases, 21 July, 20 August & 26 August 2026). Figure 2. Plan of the Southwest Main Sulfide Corridor, showing drill collars, reported intercepts and interpreted geology, with assays pending on holes still to report. Source. Terra Metals, ASX release 25 August 2026. Southwest significant intercepts, by ASX release Figure 3. Cross-section A to A′ through the Southwest Main Sulfide Corridor, looking north, with reported intercepts and the interpreted high-grade reefs inside the broader mineralised package. Source. Terra Metals, ASX release 25 August 2026. Figure 4. The Dante Project at project scale. The Dante Reefs Mineral Resource over the Celus, Hyperion and Oceanus reefs, the Southwest discovery, and the mapped reef trends. The company states Southwest at greater than 1,100 m of strike, around 700 m width, at least 750 m down dip, and open in multiple directions. Source. Terra Metals, ASX release 25 August 2026. Figure 5. Cross-section C to C′ through the Southwest Main Sulfide Corridor, looking north, showing the drilled extent of the mineralised package across roughly a kilometre of section. Source. Terra Metals, ASX release 25 August 2026. TABLE 3. SOUTHWEST SIGNIFICANT INTERCEPTS, BY ASX RELEASE Plate 3. Drill core from the Southwest Main Sulfide Corridor. The company describes semi-massive to massive sulphide with coarse visible chalcopyrite, pentlandite and pyrrhotite, and associated bornite and cubanite. Source. Terra Metals, Tribeca Future Facing Symposium 2026 presentation, ASX release 25 August 2026. 4 COMPARABLE DEPOSITS: HOW DANTE REEFS SITS GLOBALLY One way to judge an early-stage resource is to place it beside the deposits it most resembles. Dante Reefs is a layered mafic intrusion hosting stratiform, vanadium-bearing titaniferous-magnetite reefs that also carry copper and platinum-group metals, a polymetallic oxide-reef system (148 Mt @ 14.8% TiO₂, 0.54% V₂O₅, 0.18% Cu, 0.33 g/t PGE3). That specific combination (bulk titanium-vanadium magnetite plus precious metals, in stratiform reefs) is uncommon. The global comparables fall into two groups. The giant vanadium-titanium-magnetite (VTM) systems that define the commodity, and the rarer polymetallic magnetite reefs that also carry PGE-Au, which Dante most closely resembles. (The comparable-deposit figures below are drawn from public company and technical sources and are approximate. Grades are quoted on the basis reported by each project, whether whole-rock, in-magnetite, or in-concentrate, which is noted where it matters.) Verification note: Panzhihua's composite tonnage and its ~40% global vanadium-output share, and Largo's ~3.3% V₂O₅ concentrate grade, could not be independently corroborated in this pass and should be confirmed against each company’s own current disclosures. TABLE 4. COMPARABLE DEPOSITS: DANTE REEFS (TI-V MAGNETITE) The archetypes. Layered intrusions with V-Ti magnetite reefs Bushveld Complex, Upper Zone (South Africa). Producing. The Bushveld is the world's largest layered mafic intrusion and the reference point for this deposit class. Its Upper Zone carries four to five laterally continuous vanadiferous-titaniferous magnetitite layers (individual magnetite seams typically run 1 to 2.5 m thick, recurring across an Upper Zone sequence that spans hundreds of metres stratigraphically) with in-magnetite V₂O₅ grades of roughly 1.6% to 2%+ and vanadium resources exceeding 500 Mt of ore. Critically, the same intrusion hosts the Merensky Reef and UG2, the world's pre-eminent PGE reefs, in its Critical Zone. Bushveld therefore separates into different zones what Dante appears to combine in one horizon set. V-Ti magnetite reefs and precious-metal mineralisation in a single layered system. It is the geological template Terra's own material points to. Panzhihua, Sichuan (China). Producing. The Panzhihua layered intrusion, part of the Emeishan large igneous province, is the giant of the VTM world. The main intrusion is reported at approximately 1.33 billion tonnes at ~33% Fe, ~12% TiO₂ and ~0.3% V₂O₅, and the Pangang operations built on the district account for on the order of 40% of global vanadium output. Dante's whole-rock titanium (14.8% TiO₂) and vanadium (0.54% V₂O₅) grades stand comparison with Panzhihua's, and on vanadium exceed them, but Dante is a small fraction of the tonnage and entirely undeveloped. Maracás Menchen / Rio Jacaré (Bahia, Brazil). Producing. Largo's Maracás mine sits in the Rio Jacaré mafic-ultramafic intrusion, a sheet-like body with a reported 40 km of strike length in which mineralisation occurs as cyclic magnetitite, magnetite-pyroxenite and magnetite-gabbro layers, a reef-style architecture directly analogous to Dante's. Vanadium is hosted in titaniferous magnetite, and the operation produces a very high-grade magnetite concentrate (~3.3% V₂O₅) as one of the world's highest-grade, lowest-cost vanadium producers. It demonstrates the reef-style magnetite model in production. The closest analogue. A polymetallic magnetite reef with precious metals Skaergaard / Platinova Reef (East Greenland). Resource / scoping. Of all the comparables, Skaergaard is the closest to Dante in character. It is a layered mafic intrusion whose Platinova Reef is a stratiform horizon carrying palladium, platinum and gold together with iron-titanium oxide and base-metal sulphides, with vanadium, titanium, iron and gallium in the intermediate levels. Precious metals and a Ti-V-Fe oxide budget in the same reef, which is exactly the polymetallic combination Dante shows. It holds a historical JORC resource (Platina, 2013) and a 2019 scoping study, with more recent resource-definition drilling. A 2026 technical report tied to the project's current owner has floated a headline valuation of US$68 billion, a figure so far outside anything else in this table that Samso flags it as unverified and treats it with strong scepticism pending independent review. The 2013 JORC resource and 2019 scoping study remain the only figures relied on here. Like Dante it remains pre-development. Unlike Dante its precious-metal endowment is Pd-Au led rather than a PGE3 basket. The development-stage benchmarks. Where a Dante-style project must go Mount Peake (Northern Territory, Australia). DFS / development. TNG's Mount Peake is a flat-lying V-Ti-Fe orebody in a layered gabbro, with a JORC resource of about 160 Mt at ~0.28% V₂O₅, 5.3% TiO₂ and 23% Fe and a completed definitive feasibility study for an integrated mine-and-processing operation. It is the most advanced Australian analogue and a marker for the metallurgical and processing path a titanium-vanadium project must complete, work Dante has not yet begun. Mont Sorcier (Quebec, Canada). PEA / feasibility. Voyager Metals' Mont Sorcier is a large layered magnetite body with indicated resources of about 679 Mt at ~0.20% V₂O₅ (plus roughly 547 Mt inferred), notable for very low titanium, which simplifies vanadium extraction. It is advancing toward feasibility, and illustrates both the scale these systems can reach and the fact that titanium content, high at Dante, materially affects processing. What the comparison says about Dante Two conclusions follow. On geology and grade, Dante Reefs is a credible member of this class. Layered-intrusion magnetite reefs with titanium and vanadium grades that compare with producing VTM operations, and a polymetallic Cu-PGE character that, among the comparables, only Bushveld and Skaergaard share. On stage, Dante is the least advanced of every deposit listed here. Bushveld, Panzhihua and Maracás are in production. Mount Peake and Mont Sorcier are at DFS or feasibility. Skaergaard has a resource and a scoping study. Dante Reefs remains pre-scoping. The geology places Dante in good company. The development timeline is a reminder of how far a bulk, remote titanium-vanadium project still has to travel before the resource becomes cash, which is a large part of why the market's attention has moved to Southwest. 5 THE NEWSFLOW SINCE THE BROKER REPORT The broker's valuation was published on 6 May 2026. To place the recent releases in context, the timeline below opens with the three earlier events that set up the Southwest story, then covers the price-sensitive drilling and corporate releases Terra has reported since. Over this period the company has refreshed its board, added to its development team, and seen turnover in its share register. Before the broker report: Jan 26. SW5. 52.97 g/t Feb 26. A$85m raise Apr 26. SW6. 172.4 m Since the broker report (6 May 2026): 10 Jun. First 2026 assays. SWDD011 100 m @ 1.06 g/t, SWDD009, SWRD051 18 Jun. Infill. SWRC058 20 m @ 2.12 g/t, SWRC033 25 Jun. Board refresh. Debra Bakker appointed Chair. Quartermaine & Purdie join as NEDs (see §8) 1 Jul. SWRD057 194 m @ 0.71 9 Jul. Development team strengthened (Hammond, Jones) 13 Jul. "Southwest Discovery Continues to Grow": SWRC064 157 m @ 0.73 g/t from 5 m (near-surface). Corridor widened to ~700 m 21 Jul. First deep diamond assays. SWRD051 448 m @ 0.87 g/t PGE3 from 102 m 30 Jul. June 2026 quarterly report and an Appendix 2A securities-quotation amendment. The quarterly reports A$75.1m cash at 30 June 2026 20 Aug. 150 m strike extension, corridor to >1,100 m, still open 26 Aug. Further assays confirm continued mineralisation growth (SWRD057 392 m combined). >56,000 m drilled at Southwest, ~32,000 m of assays pending CEO Thomas Line described the 1 July results as confirming a large, continuous system that ended in mineralisation in every reported hole, "outside the existing Dante Mineral Resource." (TM1 ASX release, 1 July 2026.) On 13 July 2026, the company reported further RC results confirming the corridor is continuous along strike and down-dip and now approximately 700 m wide. SWRC064 intersected 157 m at 0.73 g/t PGE3 from 5 m, demonstrating the system extends to near-surface and, per the company, supporting the potential for a low strip-ratio starter pit. Management stated the higher-grade internal reef horizons compare favourably in grade and thickness with globally significant PGM reefs such as the Merensky Reef and UG2 (TM1 ASX release, 13 July 2026). The share price in context The re-rating has not been a straight line, and the sequence matters. From roughly A$0.02 in April 2025, Terra ran to almost A$0.20 through September to November 2025 as the market responded to the Dante story and the maiden Dante Reefs resource, before fading back into the A$0.10 range in the months before Southwest. The Southwest results in early 2026 then drove a second, sharper leg, with the stock spiking to its 52-week high of A$0.47. Since that post-discovery peak the share price fell hard through to mid-July 2026 (down ~31% in the month to 13 July), then traded roughly flat through six weeks of further positive Southwest newsflow, sitting at A$0.235 as at 27 August 2026 (ASX), still up sharply over twelve months, though the exact multiple depends on the comparison date. (Levels above are approximate, read from the price chart rather than from daily closes.) Does the market reward the drilling news? An event-by-event look The two biggest single-day moves in this stock both came before the broker report, and both make sense on their own terms. The market was closed for Australia Day on 26 January 2026, so when the SW5 discovery result hit the wire that morning, the reaction landed on the next session. The stock opened sharply higher and closed at A$0.35, up from A$0.21 the prior close, a 67% move on 27.7 million shares traded (against typical daily volume in the low millions). The February capital raise shows a different, more mundane pattern. The stock was halted, with zero recorded volume on 25 and 26 February 2026, and resumed trading flat on 27 February at A$0.41, in line with standard ASX practice around a placement rather than any sign of market disapproval. From the broker report onward, TM1's own release dates are precise, and so is the daily price record, which makes a cleaner test possible. Compare the closing price on each announcement day against the prior session's close. TABLE 5. ANNOUNCEMENT-DAY PRICE REACTION, 6 MAY TO 26 AUG 2026 Six of the ten dated announcements closed lower than the prior session, including three (10 June, 1 July, 13 July) that this workup's own newsflow section frames as substantive results. Only four closed higher, and the two clearest positive reactions (6 May, 21 July) both resolved a specific open question (a valuation existing, and depth confirmed) rather than simply adding another set of assays to an already-long list. Extending the window makes the pattern more decisive, not less. Of the nine events with a full five trading days now on the record, six sat lower against the prior close by then than on the announcement day itself, including 25 June and 9 July, whose same-day moves looked mild. Only 21 July and 20 August still showed a net gain at five trading days out. Ten trading days out is starker again. Of the eight events old enough to measure, seven sat below the pre-announcement price, 21 July the lone exception, and even 6 May, the single best same-day reaction in the table at +18.2%, had round-tripped to a small loss (−1.8%) by then. A positive open on drilling news has not, on this record, been a reliable sign that the market kept its conviction a fortnight later. The broader trend across the same window is a market cap that has drifted down from the A$0.325 close on the broker report to A$0.235 now, even as the drilling itself, by TM1's own numbers, has done nothing but grow the corridor and confirm grade. Two readings are both consistent with that pattern and neither can be ruled out from price action alone. The market priced most of the Southwest story into the stock during the January to February run and is now waiting for the resource itself before paying for further increments of it. Or the market is genuinely more sceptical of this discovery converting to a mineable resource than the geological results alone would suggest. This is a price correlation across three fixed horizons, not a controlled study. It does not net out broader small-cap resource sector moves, sentiment already built in ahead of a release, or the standard "buy the rumour, sell the fact" pattern common to speculative explorers generally, so treat the trend as suggestive of how the market has been behaving, not as a verdict on whether that behaviour is right. Chart 2. Samso Research, from Yahoo Finance daily closes, built with Samso Chart Studio. Milestone callouts and the "highest close" label added by Samso from the same daily-close series (the ASX-quoted 52-week intraday high of A$0.47 is not shown, since this is a closing-price chart). Prepared 27 August 2026. Two points follow. First, the 2025 run and subsequent fade suggest the market's conviction in the titanium-vanadium story was not durable. Second, and less comfortable for the bulls. The Southwest discovery has not, so far, sustained a higher price. The market drove the stock up on the early results and has been selling it down since, even as the drilling has continued to deliver. Strong geology and a falling share price are both true at once (the market is repricing sentiment faster than the geology has changed), and this report holds both, rather than citing whichever figure suits the case. "Our focus is now firmly on converting this expanding mineralised system into a maiden Mineral Resource Estimate, supported by an intensive drilling campaign with five rigs currently operating." Thomas Line, Managing Director and Chief Executive Officer, TM1 ASX release, 26 August 2026. That is the company naming its own settling event. A maiden Mineral Resource Estimate is the first time a tonnage and a grade are put on Southwest under the JORC Code, signed off by a Competent Person. The table above shows the market has not paid for the strike extensions that came before it. On that record, the resource is the release that matters, and the drilling updates between now and then may move the price very little either way. 6 THE BROKER'S NUMBER, AND THE POTENTIAL UPSIDE The broker's valuation is built as follows. To value Southwest, the model assumes a mineralised envelope (strike, width, thickness, density and a fill factor), arrives at a conceptual 57.1 Mt, applies assumed recoveries, then probability-weights the result into a figure of A$0.84 a share (East Coast Research, 6 May 2026). Every input is an assumption, because there is no Southwest resource yet. It sits expressly outside the formal estimate, and the company's own JORC table states the drill spacing is "not currently sufficient to establish… continuity" for a resource. Because each input is an assumption, the resulting figure carries the uncertainty of those assumptions, however precise it appears. The potential upside Current market cap is A$237.10 million (ASX, 27 August 2026). If Southwest converts from an open corridor into a maiden resource of scale, a company at this stage can re-rate by a multiple of its market cap. A five-fold re-rating is possible in a favourable outcome, consistent with what significant magmatic-sulfide discoveries in Western Australia have achieved previously. The clearest comparable is Chalice Mining's Julimar/Gonneville discovery, the benchmark for this style, which Terra's own releases cite, since Nebo-Babel. Illustrative range only, not a forecast or a valuation. From ~A$237m (ASX, 27 August 2026): 2x → 5x → 10x, and the downside case of a halving. Each step assumes Southwest becomes a defined resource. The counterweight For every Chalice, the ASX has produced numerous "next Julimars" that returned a strong hole, raised capital on it, and never converted the result into a defined resource. Most did not multiply. Most returned toward their cash backing. A market cap that can rise five-fold can also halve, and Terra has already demonstrated the downside on a smaller scale, falling ~31% in a month despite positive results. The multiple describes the potential range for this class of discovery, not the outcome for this one. Chalice itself is the cautionary case. It peaked near A$10.12 in November 2021 on its maiden resource and has since fallen more than 60% (Section 12). 7 WHAT WOULD CHANGE OUR MIND Figure 6. Bouguer gravity, theta derivative, over Southwest. The company reads the gravity features as aligning with the mineralised ultramafic unit and has interpreted further anomalies along strike as possible extensions. Drill approvals have been obtained to test them. Source. Terra Metals, Tribeca Future Facing Symposium 2026 presentation, ASX release 25 August 2026. Plate 4. Heritage survey work on the Dante Project. Terra Metals reports a Native Title Agreement across the project and ongoing engagement with Traditional Owners through the Ngaanyatjarra Land Council, with progressive heritage surveys supporting exploration access. Source. Terra Metals, Tribeca Future Facing Symposium 2026 presentation, ASX release 25 August 2026. The key milestone. A maiden Southwest resource. The development that would resolve the question is Terra converting Southwest's open corridor into a JORC Mineral Resource Estimate, which the company is targeting for late 2026. That is the point at which a series of drill results becomes a defined deposit with a stated size and confidence category. Catalysts up. The first deep diamond assays (up to ~918 m, the first real test of the system at depth) returned in July and August 2026 and held grade, with SWRD051 confirming 448 m at 0.87 g/t PGE3 from 102 m. Roughly 32,000 m of further assays remain pending as at 26 August 2026, and continued step-out and infill holding grade and width, a maiden Southwest MRE in late 2026, metallurgy confirming a saleable concentrate, and a scoping study that would, for the first time, put an economic study around the project are all still ahead. The February 2026 A$85m placement gives the balance sheet to fund the drilling rather than dilute into it. The June 2026 quarterly reports A$75.1m in cash at 30 June 2026, and a subsequent Appendix 2A amendment (30 July 2026) lifted the share count to about 1.03 billion. The bear case Southwest does not yet have a formal resource, which is a stage of the process rather than a verdict on the discovery. The company's own JORC table states the drill spacing is not yet sufficient to establish continuity, and the headline thick intervals are RC pre-collars with the deeper diamond assays still pending. Infill drilling can firm up a resource estimate or reduce it just as readily. Either outcome is a normal part of converting a discovery into a defined deposit, and on its own it says nothing about whether the mineralisation is there, only that the confirming work is not finished. Where the genuine risk sits is elsewhere. Dante Reefs is bulk, low-grade Ti-V in the remote West Musgrave, presenting significant infrastructure and capital challenges. With about 1.03 billion shares on issue (ASX, 27 August 2026, up from ~973 million at 13 July 2026 following a 30 July 2026 Appendix 2A securities-quotation amendment) and no cash flow, Terra will most likely raise capital again. Further dilution is close to certain, with only the price uncertain. PGE and base-metal prices can weaken and reduce sentiment with them. The market has also demonstrated it will sell the stock heavily. None of this makes the discovery less real, but it is genuine risk to weigh against it. 8 MANAGEMENT The board and management have been materially strengthened for the shift from explorer to developer, the part of the story worth weighing most. The 25 June 2026 board refresh (effective 29 June) brought three appointments, with Ian Middlemas (Chairman) and Ben Cleary stepping down (TM1 ASX release, 25 June 2026): Debra Bakker. Independent Non-Executive Chair. A mining financier and director with ASX-100 governance and capital-markets depth. Currently on the boards of Yancoal (ASX: YAL) and Ten Sixty Four, and a director of IGO (ASX: IGO) for a decade until recently (the owner of the Nova-Bollinger mine discussed in Section 12), and formerly Head of Metals & Mining Origination at Commonwealth Bank. The capital-markets experience a company that must fund development will need. Jeff Quartermaine. Independent Non-Executive Director. Recently retired MD & CEO of Perseus Mining (ASX: PRU) after nearly 13 years, having built it from a junior, single-asset gold explorer into an ASX-100, multi-mine, pan-African gold producer (more than 500,000 oz per year from 2022). The DNA question answered in the right direction. Someone who has taken a company from the ground to production. Dr Jennifer Purdie. Independent Non-Executive Director. Former BHP Asset President of Olympic Dam (a large, complex polymetallic operation), with senior roles at Rio Tinto, Jemena and Adani's Australian renewables business. Directly relevant operational and technical depth for a polymetallic project. They join CEO & Managing Director Thomas Line and Independent Non-Executive Director Haydn Smith. On paper it is a marked step up in board calibre for a company at this stage. A capital-markets chair, a proven mine-builder and a heavy-industry operator, assembled expressly for the explorer-to-developer transition. Terra then strengthened its development team on 9 July 2026 (TM1 ASX release, "TM1 Strengthens Development Team"). - Ben Hammond. Chief Development Officer. 22+ years across mine operations, project development and corporate leadership. Most recently Technical Director at TSX-listed WSP Global, advising development banks, commercial banks and mining funds on project finance and acquisition due diligence. The profile of someone preparing a large exploration position for development-funding conversations. - Barry Jones. General Manager – Exploration & Operations (and the named Competent Person on the 13 July 2026 results). 25+ years international exploration, greenfields to resource definition. A recognised specialist in layered-mafic PGE-Cu-Ni systems, with nine years at Anglo American Platinum on the Bushveld Complex supporting the Mogalakwena deposit, directly relevant to Dante's Bushveld-style geology. With a register of institutional holders conducting their own due diligence, the management and ownership of the company compare well against most peers at this market cap. Capability is not certainty, and a strong team does not convert a corridor into a resource on its own, but for an early-stage investment, this is the calibre of leadership an investor would want in place. 9 THE MARKET: WHAT IS TM1 SELLING INTO? To understand the re-rating, separate what Terra has defined from what Terra is being priced for. What the market is not paying for. Titanium and vanadium. The defined 148 Mt Dante Reefs resource is, by contained metal, principally a titanium-vanadium deposit. Both sit in subdued markets. Titanium feedstock demand is muted with limited near-term upside, and vanadium has been weak through 2024 to 2025 on soft steel demand, with a recovery not widely expected until late 2026 and beyond as vanadium-redox-flow-battery demand builds. Combined with the bulk, low-grade and remote nature of the deposit, this is not the asset that has re-rated the stock. (Commodity-market conditions as at mid-2026.) What the market is paying for. The Southwest basket. Southwest is a platinum-group-metal (platinum, palladium and gold), copper, nickel and cobalt sulphide system. That basket maps onto two of the stronger thematics in mining today: Platinum-group metals. The medium-term PGM case is constructive on the supply side. Platinum sits in structural deficit and forecasters have projected higher average prices for 2026 (Metals Focus put platinum's 2026 average near US$2,190/oz. Bank of America more bullish still). The spot market, however, is softer than those forecasts. As at 29 July 2026 platinum is about US$1,605/oz and palladium about US$1,263/oz, platinum up around 2.5% and palladium up around 4% over the preceding month (Trading Economics, 29 July 2026). The structural story is favourable. The current price still sits well below the higher 2026 forecast average, a distinction that matters for any project whose economics assume higher prices (see the Julimar lesson in Section 12). Copper, nickel and cobalt. These are the electrification and critical-minerals metals. Copper for grids and electrification broadly, nickel and cobalt for batteries and stainless steel. They carry a structural-demand narrative that a titanium-vanadium resource does not. Why the discovery adds value. The re-rating is not the market suddenly valuing titanium. It is the market beginning to price the possibility that Terra has found a new high-grade PGE-Cu-Ni-Co system (in the same magmatic-sulphide family as Chalice's Julimar and BHP's Nebo-Babel) in a rising PGM market. That is the prize investors are looking at, and it is why the Southwest holes, not the Dante resource, moved the share price. The question that remains. A strong end-market does not make a deposit. Terra is now priced for exposure to metals that are wanted, but it has no Southwest resource, no economic study and no confirmed metallurgy or recoveries, the very things that turn a favourable market into cash. The market backdrop explains why investors are interested. It does not, on its own, justify any particular price. That distinction is the heart of the Watch & DYOR call. One further force sits behind several of these prices (China's dominance of critical-metal supply, refining and demand), which Section 14 takes up in detail. 10 HOW GOOD IS DANTE REEFS: AND IS IT GOOD ENOUGH? This section answers one question. On the facts that can be sourced today, is the Dante Reefs resource good enough to support the valuation the market is implying? It uses only sourced information. Where the facts do not allow a conclusion, that is stated as an open issue rather than filled with an assumption. How good is Dante Reefs? On the sourced record, Dante Reefs is a real, JORC-classified resource of genuine size and commodity diversity: Size (sourced). 148 Mt grading 14.8% TiO₂, 0.54% V₂O₅, 0.18% Cu and 0.33 g/t PGE3 (1.38% CuEq), for contained metal of approximately 22 Mt TiO₂, 800 kt V₂O₅, 270 kt copper and 1.6 Moz PGE3 (TM1 ASX release, 11 August 2025, reproduced 13 July 2026, Table 1). Grade in context (sourced). On the comparison in Section 4, Dante's whole-rock titanium (14.8% TiO₂) and vanadium (0.54% V₂O₅) grades stand with, or above, producing vanadium-titanium-magnetite operations, for example Panzhihua (~12% TiO₂, ~0.3% V₂O₅) and Mount Peake (5.3% TiO₂, ~0.28% V₂O₅), and Dante carries copper and PGE credits that a straight VTM deposit does not. Classification (sourced). The resource is reported under JORC 2012. Of the 148 Mt, 38 Mt (26%) is Indicated and 110 Mt (74%) is Inferred (TM1, 11 August 2025). Those are the facts that make Dante "good": scale, competitive oxide grades, and a polymetallic basket. How good could it be? The sourced record shows expansion potential, but not its size: Multiple reef horizons (sourced). The Jameson Layered Intrusion is interpreted to host at least 11 PGM-Cu reef horizons, and the current estimate covers only three deposits. Crius, Hyperion and Oceanus (TM1 ASX release, 13 July 2026, geology section and Table 1). Strike length not yet in the estimate (sourced). Company figures label reef trends well beyond the resource, for example Reef 1 (~17 km strike), Reef 2 (~12 km) and Reef 5 (~7 km), of which only portions are captured in the 148 Mt estimate (TM1 ASX release, 13 July 2026, Figure 3). Those facts establish that the resource is open to growth. Open issue. The amount of additional tonnes or metal that could be added cannot be stated from sourced facts (no expanded estimate has been reported), so any specific "upside" figure would be an assumption, and none is offered here. What could go wrong? Again, only what the sources support: Confidence (sourced). 74% of the resource sits in the lower-confidence Inferred category (TM1, 11 August 2025). Inferred material may or may not convert to higher confidence with further drilling. No economic study (sourced). There is no scoping study, no pre-feasibility study and no feasibility study on Dante Reefs. The company's stated economic-study pathway is tied to a maiden Southwest resource targeted for late 2026 (TM1 ASX releases, 1 & 13 July 2026). Without an economic study, capital cost, operating cost and margin are undetermined. Metallurgy (now partly answered, see Section 11). Terra's Phase 1 testwork (25 March 2025) produced copper-gold-PGM, vanadium and titanium concentrates by simple magnetic separation and flotation, with strong copper (95.8%) and vanadium (90.9%) recoveries and workable PGM (74.4%) and titanium (65.6%) recoveries. That de-risks first-principles processability. What remains open is a confirmed saleable titanium grade, downstream conversion to final products, pilot-scale confirmation and economics, assessed in full in Section 11. Commodity backdrop (sourced). The principal contained metals by tonnage are titanium and vanadium, and both sit in subdued markets (see Section 9). Weak prices for the primary products weigh on the economics of a bulk, low-grade deposit. Location (sourced). Dante is in the remote West Musgrave region of Western Australia (TM1 ASX release, 13 July 2026, "About the Dante Project"). Remoteness is a fact. Its cost implications depend on an economic study that does not yet exist (open issue). Financing (sourced). The company has about 1.03 billion shares on issue (ASX, 27 August 2026) and no cash flow. Further capital raising is a realistic consideration, though its timing and price cannot be stated as fact. Is the grade and resource good enough? Split into what can and cannot be concluded from sourced facts: On grade, the facts are supportive. Dante's titanium and vanadium grades compare with, or exceed, producing VTM operations, and the deposit adds copper and PGE credits (Section 4). On grade alone, there is no sourced basis to call the resource sub-standard for its class. On economic adequacy, no conclusion is possible. Whether the grade and resource are "good enough" to be mined economically cannot be determined from the sourced record, because the two things that would settle it, an economic study (scoping/PFS/feasibility) and reviewed metallurgical recoveries, do not exist in the material available. This is an open issue. On the current valuation, the sourced facts point elsewhere. The one published valuation, East Coast Research's A$0.84 fair value (6 May 2026), is expressly modelled on Southwest, using a conceptual 57.1 Mt envelope and assumed recoveries, and the company's own JORC table states Southwest's drill spacing is "not currently sufficient to establish continuity" for a resource. On the record, the implied valuation is built on an undefined Southwest, not on the defined Dante Reefs. The open issues, stated plainly For the reader, these are the questions the facts cannot yet answer, and which will decide whether TM1 grows into its valuation: Can Dante Reefs be shown to be economic? No scoping, pre-feasibility or feasibility study exists. 1. Are the Phase 1 metallurgical results confirmed and scalable? Phase 1 (25 March 2025) produced three concentrates with encouraging recoveries (Section 11). A confirmed saleable titanium grade, downstream product conversion, Phase 2/pilot results and economics remain to be shown. How much can the resource grow? Multiple reef horizons and long strike are documented, but no expanded estimate has been reported. Will Southwest convert to a JORC resource? The company states drill spacing is not yet sufficient to establish continuity. Until these are resolved with sourced data, the position the facts support is this. Dante Reefs is a good resource of its type whose economic adequacy is unproven, and the company's current valuation rests on Southwest rather than on Dante Reefs. That is the basis of the Watch & DYOR call. 11 THE METALLURGY OF DANTE REEFS: CAN THESE ORES BE PROCESSED? Metallurgy (can the titanium, vanadium, copper and platinum-group metals actually be recovered into saleable products?) is the question that most conditions Dante's value. This section does two things. It sets out what the published scientific literature says about processing ores of this type (vanadiferous titanomagnetite with reef-hosted precious metals), and then it tests those findings against Terra's own Phase 1 metallurgical results (ASX announcement, 25 March 2025). The merit. These ores are a proven global source of vanadium and iron Vanadiferous titanomagnetite is not exotic. It is the dominant global source of primary vanadium, produced either directly from titanomagnetite concentrates or from the slag left after smelting them, from deposits in China, South Africa and Russia ("The extraction of vanadium from titanomagnetites and other sources," Minerals Engineering). The established industrial route is salt (soda) roasting followed by water leaching, which converts vanadium locked in the magnetite lattice into water-soluble sodium vanadates ("Vanadium, Titanium, and Iron Extraction from Titanomagnetite Ore by Salt Roasting…," Separations (MDPI), 2025). Panzhihua and the Bushveld (two of the comparables in Section 4) run at commercial scale on exactly this basis. So the base case is not "can it be done". It demonstrably can. The challenge. Titanium is the hard part, and Dante is titanium-rich The difficulty in this ore class is titanium, and Dante Reefs is a high-titanium deposit (14.8% TiO₂). The literature is consistent on why that matters: In conventional blast-furnace smelting, TiO₂ reports to the slag, and at 20% to 25% TiO₂ the slag becomes viscous and the furnace difficult to operate. Keeping slag TiO₂ below ~20% is preferred for stable operation ("Development of Intensified Technologies of Vanadium-Bearing Titanomagnetite Smelting," J. Iron and Steel Research International. "Blast furnace ironmaking process with super-high TiO₂ in the slag," Int. J. Minerals, Metallurgy and Materials). Almost all the titanium ends up in that slag, but as low-grade (10% to 25% TiO₂), mineralogically complex material that is difficult to recover, so titanium is often a lost by-product, and a share of the vanadium is lost to the slag as well (same sources). This is precisely why a low-titanium VTM is considered easier to process, the reason Mont Sorcier (Section 4) markets its low-Ti character as an advantage, and why high-Ti projects tend to avoid the blast furnace altogether. The response. Alternative flowsheets exist, but add complexity The literature shows the industry's answer is to bypass conventional smelting with roast-leach and hydrometallurgical routes, several of which achieve high recoveries at research or pilot scale: Hydrogen-based direct reduction of raw VTM has reported ~92.6% iron metallisation ("Preparation of titanium mineral from vanadium titanomagnetite concentrates by hydrogen reduction and acid leaching," ScienceDirect). Partial carbon reduction with mild acid leaching can recover a TiO₂-enriched product ("Recovery of TiO₂-enriched material from vanadium titano-magnetite concentrates…," Hydrometallurgy). Deep-eutectic-solvent leaching (choline chloride/oxalic acid) has reported ~97.6% V, ~76.1% Ti and ~68.8% Fe leaching efficiencies ("…Salt Roasting and 21st-Century Solvents," Separations (MDPI), 2025). These are encouraging, but the qualifier matters. High recoveries in a laboratory or pilot do not translate automatically into a commercial flowsheet at acceptable cost, and each added processing step is capital and operating cost that a bulk, low-grade, remote deposit can least afford. The precious-metal question. PGE in oxide, not sulphide Dante's value is not only Ti-V. It carries copper and a platinum-group basket, and here the literature raises a separate, specific caution. Conventional froth flotation recovers upwards of 85% of platinum from pristine sulphide ores, but where PGE are hosted in or associated with oxides rather than clean sulphides, recoveries by conventional flotation fall sharply (historically below 50%) because altered/oxide surfaces do not respond to standard sulphide collectors ("Potential Processing Routes for Recovery of PGMs from Southern African Oxidized PGM Ores: A Review," J. Sustainable Metallurgy, 2017). Specialised reagent schemes have lifted recoveries, one study improving from ~55% to ~75% 3E (Pt-Pd-Au) with a hydroxamate oxide collector ("PGM extraction from oxidized ores using flotation and leaching," SAIMM Journal), but only with tailored metallurgy. Because Dante's PGE sits within magnetite-rich reefs rather than a classic sulphide ore, the mode of PGE occurrence and its recoverability is exactly the kind of question project-specific testwork must answer. Dante's own results. How they test the literature Terra's Phase 1 testwork, on representative composite samples from Dante Reefs diamond drilling, used simple magnetic separation and flotation to produce three separate concentrates (TM1 ASX announcement, 25 March 2025. figures as reported by the company): Copper-gold-PGM sulphide concentrate. 28.0% Cu, 17 g/t Au, 21.4 g/t PGM, at recoveries of 95.8% Cu, 75.8% Au and 74.4% PGM. Vanadium-magnetite concentrate. 1.81% V₂O₅ at 90.9% recovery, described by the company as suitable for high-purity vanadium-pentoxide flake. Titanium-ilmenite concentrate. 40% TiO₂ at 65.6% recovery, produced by magnetic separation alone. Chart 3. Samso Research, from company-reported Phase 1 testwork (TM1 ASX, 25 Mar 2025). Set against the literature, three things stand out: The titanium hurdle is side-stepped, not confronted by smelting. Dante's flowsheet does not go near a blast furnace. It recovers a titanium-ilmenite concentrate by physical (magnetic) separation, which avoids the high-TiO₂ slag problem the literature flags. The qualifier is grade and recovery. 40% TiO₂ at 65.6% recovery is a starting point, not a finished answer. The company states it needs ~46% TiO₂ to be saleable and may upgrade by flotation (the route used in China and Brazil). Titanium recovery is plausible but not yet at a confirmed saleable spec. The precious-metal caution largely does not apply here. The literature's warning concerned PGE locked in oxides, which flotation struggles to recover (historically <50%). Dante's PGM instead reports to a sulphide concentrate at 74.4% recovery, alongside 95.8% copper. The platinum-group metals are sulphide-associated and float, rather than being oxide-locked. That is materially better than the oxidised-ore worst case, though still short of the >85% achievable from the cleanest sulphide ores. Terra's principal metallurgist has stated the mineralogy has "favourable properties compared to some other known deposits in production." Vanadium recovery is strong at the concentrate stage. 90.9% recovery into a 1.81% V₂O₅ magnetite concentrate is a good result. The outstanding step (common to all these ores) is the downstream conversion of that concentrate into a vanadium-pentoxide product, whose cost and recovery Phase 1 does not address. What this means for Dante The metallurgy open issue is now materially better-informed. On Terra's own Phase 1 evidence, Dante's ore is not a metallurgical dead-end. A simple magnetic-separation-and-flotation flowsheet has produced three concentrates, the PGE behave as recoverable sulphides rather than oxide-locked refractory material, and vanadium recovery is high. That is a genuine de-risking of the single biggest question hanging over the resource. What remains open is scale and economics, not first-principles processability: Phase 1 is bench-scale on composite samples. Phase 2 optimisation is underway but not yet reported, and there is no pilot or commercial-scale confirmation. The titanium concentrate is not yet at a confirmed saleable grade (company target ~46% TiO₂). Downstream conversion to final products (V₂O₅ flake, ilmenite payability, copper-PGM smelter terms) is not costed, and no economic study exists. The figures above are Phase 1 results reported by the company. Confirm them against the primary 25 March 2025 ASX release. The honest revision to Section 10 is therefore this. Metallurgy is no longer an unknown. It is an encouraging but unfinished result. Dante has shown its ore can be turned into concentrates by simple means. It has not yet shown, with an economic study, that doing so is profitable at the scale a bulk, remote, low-grade deposit requires. 12 COMPARABLE DEPOSITS: SOUTHWEST (Ni-Cu-PGE SULPHIDE) Southwest belongs to a different (and, for investors, more familiar) peer group than Dante Reefs. Magmatic nickel-copper-platinum-group-element sulphide systems hosted in mafic-ultramafic intrusions. This is the class that produces the discoveries the market rewards, and the one Terra's own releases benchmark against. (The comparable-deposit figures in the table below are taken from each company's own public disclosures and independent technical reports, and are approximate.) Verification note: the Nebo-Babel grade order and the Platreef/Flatreef resource figures in this table have been corrected against each company’s current public disclosures. TABLE 6. COMPARABLE DEPOSITS: SOUTHWEST (NI-CU-PGE SULPHIDE) The benchmark Terra cites. Gonneville / Julimar (Chalice). Chalice’s Julimar discovery in 2020 is the reference point for this style in Australia and the comparison Terra’s own releases invoke. Its Gonneville deposit carries a resource of 560 Mt at 0.54% nickel-equivalent (1.7 g/t palladium-equivalent), roughly 16 Moz of 3E PGE, 860 kt nickel, 520 kt copper and 83 kt cobalt, and a pre-feasibility study (completed 8 December 2025) describing a 23-year open-pit operation with a pre-tax NPV of about A$1.4 billion (about 23% IRR). But the journey from discovery is the more instructive part of the comparison, and it is a cautionary one. Chalice re-rated to A$10.12 in November 2021 on the maiden resource, then fell more than 60% from that peak. Its August 2023 scoping study alone wiped over 25% off the shares in a single day, as the market balked at the metal-price assumptions, the recovery rates and a development cost of roughly A$1.6 billion to A$2.3 billion (Stockhead. Market Index, 2023). Metallurgy was central to the pain. Chalice initially struggled to produce a saleable nickel concentrate and was punished in 2023 when it signalled its nickel circuit might need a pressure-oxidation (POX) plant. Only after roughly A$15 million and several more years of test work did it "get the float to work", reaching an ~8% nickel concentrate at 89% to 92% nickel and 85% to 88% copper recoveries, dropping the hydrometallurgical plant, and delivering a PFS the market could accept (Chalice Mining. Discovery Alert, 2025–26). The price assumptions behind that study are the other half of the lesson, and the most useful context for a reader. Chalice's 2023 study leaned on roughly US$2,000/oz palladium when the spot price was about US$1,250. Nearly three years on, as at 29 July 2026, palladium is around US$1,263/oz and platinum about US$1,605/oz (Trading Economics, 29 July 2026), both still well short of the assumed price. The assumed price never arrived. The medium-term PGM case remains constructive on the supply side (Section 9), but the gap between a study's assumed prices and the spot a reader can actually see is exactly where headline economics are made or unmade. Three lessons for Southwest follow directly. First, even the benchmark took about five years from discovery to a workable flowsheet and PFS. The road is long. Second, the thing that repeatedly hurt Chalice was metallurgy, and above all the saleability of a nickel concentrate from a polymetallic sulphide ore, precisely the class of risk flagged for Southwest in Section 13. Third, be wary of any future study (Chalice's or Terra's) whose economics lean on metal prices above today's spot. That single assumption is what turned a "world-class" scoping study into a sell-off. Gonneville is Southwest's aspiration and its warning at once. The neighbour. Nebo-Babel (BHP). Fifteen kilometres from Dante, Nebo-Babel is the district's largest defined Ni-Cu-PGE sulphide resource. 390 Mt at 0.33% Cu, 0.30% Ni and 0.23 g/t PGE3 (BHP; reproduced in TM1's 13 July 2026 release, Figure 4). It anchors the West Musgrave as a genuine Ni-Cu-PGE province and gives Southwest geological company, though its grade is well below Southwest's discovery intercepts. The high-grade producer. Nova-Bollinger (IGO). Nova, in the Fraser Range, is the modern Australian benchmark for a high-grade magmatic sulphide mine. 11.4 Mt at 2.4% Ni, 1.0% Cu and 0.08% Co, in production since 2015 and now approaching end of life. It shows that a compact, high-grade sulphide body can be a profitable mine, and that grade, not merely size, is what makes these deposits work. The thickness analogue. Platreef (Ivanhoe). Terra's 13 July release explicitly likens Southwest's thickness to Platreef, the unusually thick PGE-Ni-Cu-Au reef on the Bushveld's northern limb (Flatreef. ~93 Moz precious metals Indicated, ~144 Moz Inferred, per a November 2025 update), which began producing concentrate in November 2025. Platreef matters because most PGE reefs are narrow (0.1 m to 2 m). Platreef and, on Terra's account, Southwest are the exceptions. Broad, potentially bulk-mineable thicknesses. That thickness is the specific feature Terra is selling. What the comparison says about Southwest. On grade and style, Southwest's discovery intercepts (35 m @ 2.90 g/t PGE3, peaks to 52.97 g/t. SWDD006 172.4 m @ 1.11 g/t) sit at the higher-grade, thicker end of this peer group, genuinely encouraging. But every deposit above has something Southwest does not. A defined resource. Gonneville has 560 Mt and a PFS. Nova and Platreef are in production. Nebo-Babel is a defined 390 Mt. Southwest has drill intercepts over a large footprint and no estimate. The peer group confirms the prize is real, and for the best of them very large. It also shows Southwest is at the very start of the road those deposits have travelled. Chart 5. Samso Research, spanning the peer groups in Sections 4 and 12. 13 THE METALLURGY OF SOUTHWEST: A MORE CONVENTIONAL ROUTE, WITH ONE CATCH The metallurgy of Southwest is, on the science, a more familiar and better-understood problem than Dante Reefs, with one specific caveat rooted in its host rock. As in Section 11, every statement here is drawn from peer-reviewed scientific literature, cited so the comparison rests on published fact rather than assertion. No Southwest-specific metallurgical testwork has been reported (Terra's only released testwork, 25 March 2025, was on Dante Reefs), so this is type-literature context. Southwest's own answer remains an open issue. The merit. Sulphide-hosted Ni-Cu-PGE is the routinely-processed case. Southwest is a magmatic nickel-copper-PGE sulphide system. That is the ore type the industry processes most routinely. Conventional froth flotation produces nickel and copper sulphide concentrates, and the platinum-group metals report with the sulphides, the basis of production at Sudbury, Norilsk, Nova-Bollinger and the Bushveld. Where PGE occur in fresh sulphides, platinum flotation recoveries above ~85% are achievable, in contrast to the below-50% recoveries from oxidised ores noted in Section 11 ("Potential Processing Routes for Recovery of PGMs from Southern African Oxidized PGM Ores: A Review," J. Sustainable Metallurgy, 2017). In principle, Southwest's sulphide mineralisation is the favourable metallurgical case, the same reason Dante Reefs' own PGE floated well once they were sulphide-hosted (Section 11). The catch. An ultramafic host means magnesium-silicate gangue. Terra describes Southwest as hosted in a "newly recognised ultramafic package" (TM1 ASX release, 13 July 2026). Ultramafic-hosted nickel sulphides carry a well-documented processing problem. Magnesium-silicate gangue minerals (serpentine, talc and chlorite) which in many such ores make up the majority of the rock and are intergrown with the nickel-bearing pentlandite ("Review on the Challenges of Magnesium Removal in Nickel Sulfide Ore Flotation and Advances in Serpentinite Depressor," Minerals (MDPI), 2024). The literature is specific about why this matters: Talc floats naturally and reports to the concentrate, diluting grade and impairing selectivity ("Behaviour of talc and mica in copper ore flotation," Applied Clay Science (ScienceDirect)). Serpentine and other Mg-silicates cause slime coatings and rheological problems that reduce recovery and selectivity, and high MgO in the concentrate penalises downstream smelting ("Sustainable Flotation of Complex Copper-Nickel Sulfide Ores: A Review of Integrated Process Strategies," Mineral Processing and Extractive Metallurgy Review, 2025). Fine grinding and tailored depressant schemes are typically required to liberate the sulphides and hold the Mg-silicates back (same reviews). What this means for Southwest. The balanced, sourced position. The ore type is favourable. Sulphide-hosted Ni-Cu-PGE is the case the industry treats every day, and where the sulphides are clean the recoveries are high. The specific risk is the ultramafic host. If talc and serpentine prove abundant, concentrate grade, recovery and smelter payability can all suffer and the flowsheet becomes more demanding. Which of these dominates at Southwest cannot be stated from the type-literature, because Terra has reported no Southwest metallurgical testwork. That is the open issue. The mineralogy is favourable in kind but untested in fact, and, alongside a maiden resource, it is the most valuable thing the company could report next. The Julimar/Gonneville experience (Section 12) is the worked example of why this matters. Even that benchmark deposit (a large, high-grade magmatic sulphide system) took Chalice several years and roughly A$15 million of test work to produce a saleable nickel concentrate, having first been punished by the market when an early flowsheet looked like it would need a pressure-oxidation plant. The favourable ore type did not spare it a hard, expensive metallurgical road. There is no sourced basis to assume Southwest's will be easier until its own test work is done. 14 CHINA AND THE PRICE OF WHAT TM1 SELLS A resource is only worth what its metals sell for, and for several of TM1's metals the price is shaped (directly or at the margin) by one country. This section sets out how China's dominance of critical-metal supply, refining and demand feeds into the prices that will ultimately decide whether Dante and Southwest are economic. It is macro context, sourced, and it cuts both ways. China's grip on the critical-metals system The starting fact is concentration. China is the leading refiner of 19 of the 20 minerals commonly classed as strategically important, at an average market share of roughly 70%, and it controls an estimated 90% of global rare-earth processing, 80% of tungsten and 60% of antimony (IEA, Global Critical Minerals Outlook 2025. ORF America). Through 2025 it turned that dominance into policy. Export controls on tungsten and other metals in February, on seven heavy rare earths and magnets in April, and a further, extraterritorial tightening on rare earths, superhard materials and battery materials in October. The price effects were sharp. After China restricted antimony in August 2024, its exports to the US fell about 97% and world prices spiked sharply, with some benchmarks reporting moves of twenty-fold or more (from around US$1,400/mt to as high as US$38,000/mt within weeks), while rare-earth prices outside China ran to several times the domestic level, with magnet shortages briefly forcing some carmakers to cut output (IEA; Global Trade Alert). Several of these measures have since been eased or suspended amid US–China negotiations (Pillsbury), which only underlines how quickly the policy, and the price signal, can move. Tungsten, rare earths, gallium and antimony now trade on Chinese policy as much as on mine supply. Chart 4. Samso Research, from IEA Global Critical Minerals Outlook 2025 and ORF America. Where TM1's own metals sit in that system Each of TM1's products touches China in a different way: Titanium and vanadium. China as the dominant producer. China is the largest producer of both, anchored by the Panzhihua vanadium-titanium-magnetite district, whose Pangang operations alone account for on the order of 40% of world vanadium output (Section 4). A Western titanium-vanadium project like Dante would be selling into a market that China largely supplies and, to a degree, sets the cost of, an ex-China supply-security angle, but also direct competition against China's scale and low cost. Iron ore. China as the dominant consumer. Dante's resource carries ~23% iron. Iron ore is the clearest China-demand story of all. China takes roughly 75% of global seaborne iron ore and makes about half the world's steel, and Beijing is centralising its buying (through the China Mineral Resources Group) to gain pricing power (Reuters / GMK; Baird Maritime). When China's property and steel demand softens, the iron price falls with it, a demand risk entirely outside any miner's control. Platinum-group metals. China as the swing buyer. China mines almost no platinum and imports more than 95% of what it uses, and in 2025 it reclassified platinum as a strategic critical mineral. That shift, plus hydrogen and investment demand, drove Chinese platinum imports up sharply (one report put the jump at ~47%) and Chinese bar-and-coin investment to roughly 70% of the global total, up from about 15% three years earlier, a major reason platinum rallied to multi-year highs (NAI500. World Platinum Investment Council). Unlike titanium or iron, PGE supply is concentrated in South Africa and Russia rather than China. But Chinese demand is increasingly what sets the platinum price at the margin. Why this matters to a TM1 reader. Both ways The China factor is neither simply bullish nor bearish. It is a source of both support and risk, and a reader should hold both. The supportive case. China's willingness to use its dominance as leverage (the antimony, rare-earth and tungsten controls) is precisely why Western governments and end-users are paying up for ex-China supply of critical minerals, and why projects carrying titanium, vanadium, copper and PGE can attract a strategic premium and, in some jurisdictions, government backing. On the demand side, Chinese platinum buying is currently a genuine prop under the PGE price. The risk case. The same concentration makes prices volatile and politically driven. A metal can triple or halve on a Chinese policy announcement rather than on fundamentals. Iron ore and vanadium can sag when Chinese industrial demand weakens. And a platinum price being held up by Chinese investment buying is, by the same token, exposed if that buying reverses. These are exactly the prices any future Dante or Southwest economic study would have to assume, the link back to the Julimar lesson in Section 12 and to the no-price-target stance of this report. For the reader, the point is awareness. Several of the prices that determine whether TM1's metals are worth mining are set less in a free global market than in Beijing, as producer, consumer, refiner and policymaker at once. That is a large part of why the critical-minerals thesis exists, and a reason to treat any single assumed price with caution. 15 SAMSO CONCLUDING COMMENTS SAMSO CALL: Watch and DYOR The settling event we are watching is a maiden Southwest Mineral Resource Estimate, targeted for late 2026. Terra Metals is two stories in one company. The defined asset (the 148 Mt Dante Reefs) is a real, JORC-classified titanium-vanadium resource whose grades stand comparison with producing peers (Section 4) and whose Phase 1 metallurgy is genuinely encouraging (Section 11). But it is the least-advanced deposit in its peer group, it has no economic study, and titanium and vanadium are not the markets driving the share price. The value the market is pricing sits in the second story. Southwest, a high-grade Ni-Cu-PGE-Co sulphide discovery that, on grade and thickness, sits at the strong end of its own peer group (Section 12). WHat is missing and that may be just a matter of progress is the resource and the all important metallurgy study. That gap between what is priced and what is proven is the whole point of the call. Southwest is a substantive discovery, in favourable geology and a stable jurisdiction, with experienced management and a supportive institutional register. I do not say that of many. But the comparables are as much warning as encouragement. Even Chalice's Julimar, the benchmark Terra cites, took about five years and roughly A$15 million of metallurgical test work to turn a spectacular discovery into an acceptable study, and its share price fell more than 60% along the way (Section 12). The prices that will decide the economics (PGMs, and behind them titanium, vanadium and iron) are, for several of these metals, set as much in Beijing as in a free market (Section 14), and the one published valuation on TM1 leans on assumptions the spot market does not yet support. So the honest position is the one the facts support, and no more. This is a discovery with genuine upside and genuine uncertainty (the first deep diamond tails have now returned and held grade to 448 m, but roughly 32,000 m of further assays remain pending as at 26 August 2026) in a share price that has shown it can move 30% in a month in either direction. If Southwest converts to a resource of scale, this is the kind of company that re-rates by a multiple. If it does not, it will likely drift back toward its cash backing or lower, as this prima-dona market seem to do. Both outcomes remain possible, and nothing in the sourced record yet settles which direction is the future position. What to watch, in order. The roughly 32,000 m of Southwest assays still pending as at 26 August 2026. Phase 2 metallurgy on Dante and any first metallurgy on Southwest. And above all the maiden Southwest resource targeted for late 2026. Review the facts, take a balanced view, and reassess as the results arrive. It is not my role to tell you what to do with your money. It is my role to tell you this company warrants your attention, and exactly which unanswered questions will decide it. 16 Disclaimer The information contained on this website is the writer's personal opinion and is provided to you for information only and is not intended to or nor will it create/induce the creation of any binding legal relations. There may be a conflict of interest present with commercial arrangements with companies and/or stock held. Samso or an associate may receive a commission for funds raised. The information or opinions provided herein do not constitute investment advice, an offer or solicitation to subscribe for, purchase or sell the investment product(s) mentioned herein. It does not take into consideration, nor have any regard to your specific investment objectives, financial situation, risk profile, tax position and particular, or unique needs and constraints. Accordingly, no warranty whatsoever is given, and no liability whatsoever is accepted for any loss arising, whether directly or indirectly as a result of this information. Investments are subject to investment risks, including possible loss of the principal amount invested. The value of the product and the income from them may fall as well as rise. You may wish to seek advice from an independent financial adviser before committing to purchase or invest in the investment product(s) mentioned herein. If you choose not to do so, you should consider whether the investment product(s) mentioned herein are suitable for you. The writer will not, in any event, be liable to you for any direct/indirect or any other damages of any kind arising from or in connection with your reliance on any information in and/or materials appended herein. The information and/or materials are provided "as is" without warranty of any kind, either express or implied. In particular, no warranty regarding accuracy or fitness for a purpose is given in connection with such information and materials. The Samso Way – Seek the Research Here at Samso, we pride ourselves on delivering content for investors that is independent and informed by over three decades of experience in the industry. Our content is well-researched and is only created if I see merit in discussing the company's story. Our mission is simple. Cut through the noise and spotlight what matters. Genuine stories, grounded insights, and real opportunity. Investors can explore our four core platforms: Coffee with Samso Samso Insights Samso News Samso Research There may be numerous paths to success in investing, but the common thread among successful individuals is that they remain committed to making informed decisions. Equip yourself with the right knowledge and tools, and you will be well on your way to achieving your financial goals. The Samso Philosophy: Stay curious. Stay sharp. And remember, digging deeper always uncovers the real value. Happy Investing, and the only four-letter word you need to know is DYOR. Disclaimer This Samso Research Workup is independent research commentary for general information only. Nothing in this document is financial product advice, and it does not take into account any reader's objectives, financial situation or needs. Figures are drawn from public sources believed reliable at the stated dates but are not guaranteed; market-sensitive numbers change daily. Samso or associated parties may hold positions in, or have commercial arrangements with, companies mentioned. Seek professional advice before making investment decisions. © Samso 2026 · samso.com.au Read full Disclaimer. About Samso Samso is a trusted platform that equips dedicated investors with up-to-date industry knowledge and insights from top CEOs and thought leaders. Samso Research | www.samso.com.au | An Investor Lens on ASX-Listed Companies

Abstract Ordell Minerals is a two-year-old ASX-listed gold explorer with one asset that matters, the Barimaia Gold Project near Mount Magnet in the Murchison district of Western Australia, in which it holds 80.3 per cent. It has no Mineral Resource, no reserve and no revenue. It has told the market it expects to publish a maiden Mineral Resource Estimate for the McNabs East prospect in the December 2026 quarter. What makes the company worth a full workup is the combination of a defined near-term event and a share price that has stopped anticipating it. Ordell traded at A$1.06 within the past twelve months and closed at A$0.475 on 2 September 2026, giving a market capitalisation of about A$35.9 million against A$8.2 million of cash at 30 June 2026 and no debt. Over the same period the gold price has been high by any recent standard, at about A$6,110 an ounce on 2 September 2026, although it is not at a record and sits roughly a fifth below its January 2026 peak. The geological case is real and it has been mined. Ramelius Resources' Eridanus deposit, about 6km to the north-west, is a producing open pit hosted in granodiorite with a Mineral Resource of 24Mt at 1.7g/t gold for 1.3 million ounces and, on Ordell's own account, more than 300,000 ounces already extracted. Ordell is drilling the same host rock along more than 7km of strike. Its metallurgical testwork returned gold extractions above 95 per cent after 48 hours at a 106 micron grind, with 35 to 40 per cent of the gold recoverable by gravity alone, which removes one of the standard ways a project like this fails. What is unproven is everything that follows from that. There is no resource, so there is no size, no grade and no strip ratio. The headline high-grade intercepts are short and nuggety and will be top-cut in any resource estimate. Reported intervals are downhole lengths, not true widths. The open pit is a stated potential rather than a study. And the near-term development case depends on treating ore in a third party's plant, an arrangement nobody has announced and for which we found no evidence of a discussion. The next two quarters resolve most of this. Ordell has about 14,000m of drilling underway, of which roughly 10,000m is infill reverse circulation drilling designed specifically to confirm continuity ahead of the resource estimate. Infill drilling that reproduces wide-spaced grades is a materially different result from infill drilling that does not, and it is the single test that matters between now and December. Assume a capital raising during 2027, because the current burn rate cannot be read off the June quarter. The Samso Call is TIME TO LOOK AT TAKING A POSITION. The evidence Ordell has published is sufficient to justify a resource estimate. It is not yet sufficient to tell anyone how big that resource will be. The call describes attention, not action, and this workup contains no valuation and no price target. Contents 1 Why This Company Warrants a Full Workup 10 The Market Behind the Metal 2 The Numbers 11 Goodia and Fisher South 3 The Asset, and What Is Actually There 12 Management Weighed 4 The Drilling Record, Tested 13 The Register, the Escrow and the Share Price 5 The Science Tested 14 What Has To Go Right, and What Can Go Wrong 6 Metallurgy and the Development Path 15 What Would Change Our View 7 The Eridanus Analogue, and Where It Stops 16 Samso Concluding Comments 8 Comparables 17 Disclaimer 9 Where Ordell Sits, on Five Measures Ordell Minerals Limited - Snapshot This is a living document. Prices and market data below are as at 2 September 2026 and will have moved. Item Detail Share price A$0.475 (ASX close, 2 September 2026) 52-week range A$0.345 to A$1.060 Market capitalisation ~A$35.9 million (undiluted). Samso calculation: 75.6 million shares at A$0.475 Shares on issue 75.6 million Options and performance rights 20.4 million (unquoted) Cash A$8.2 million (30 June 2026) Debt Nil Enterprise value ~A$27.7 million. Samso arithmetic, market capitalisation less cash at 30 June 2026 Flagship asset Barimaia Gold Project, Murchison, WA (80.3%) Other assets Goodia and Fisher South, WA (100% each) Mineral Resource None reported Next catalyst Maiden Mineral Resource Estimate, December 2026 quarter Drilling underway ~14,000m, August to September 2026 Broker coverage None identified as at 2 September 2026 Substantial holders, from the holders' own Form 604 notices lodged with the ASX. Genesis Minerals Limited 7.96% (6,000,000 shares, notice of 25 May 2026). MJF and FLD Fowler and Canning View Super 5.64% (4,250,000 shares, notice of 25 May 2026). Cremorne Capital as responsible entity for the Lowell Resources Fund 7.62% (4,692,393 shares) at its last notice of 31 July 2025, which is stale and has since been diluted by an amount no notice states. Board and management about 9.0% per Ordell's June 2026 quarterly. 1 Why This Company Warrants a Full Workup Disclosure of interest. Samso Research Media House holds a small shareholding in Ordell Minerals Limited. This workup was not commissioned by Ordell and Samso has not been paid by the company to produce it. The holding is stated here, at the front, rather than only in the disclaimer at the end, so that a reader can weigh it before reading the argument rather than after. Ordell Minerals is a gold explorer with no resource, no reserve, no revenue and one project that matters. On the ordinary reading of those facts it is one of the several hundred small explorers listed on the ASX, and there is no obvious reason to spend a day on it. There are three reasons we did. The first is that Ordell has put a maiden Mineral Resource Estimate on the calendar. The company has told the market it expects to complete one for the McNabs East prospect in the December 2026 quarter. A Mineral Resource Estimate, or MRE, is the first formal count of how much metal a company believes it has in the ground, reported under the JORC Code so that it can be compared with anyone else's. Ordell has never had one. Everything the company owns today is a set of drill holes and an interpretation. In the next few months that becomes a number, and the market will either agree with the story the drill holes have been telling or it will not. The second is the neighbour. Barimaia sits next to Ramelius Resources' Mt Magnet operation in the Murchison district of Western Australia. About 6km to the north-west is Eridanus, a producing open pit with a Mineral Resource of 24Mt at 1.7g/t gold for 1.3 million ounces. Ordell states that more than 300,000 ounces have already been mined from it, a figure that appears in Ordell's quarterly rather than on Ramelius' own Eridanus page. Eridanus is not a typical Mt Magnet deposit. Most gold at Mt Magnet sits in narrow, high-grade quartz lodes. Eridanus sits in a granodiorite, which is a coarse-grained intrusive rock, and the gold is spread through a stockwork of small veins rather than concentrated in one structure. That makes it a bulk-tonnage deposit, meaning lower grade over far more rock. Ordell is drilling the same rock type, on the same district scale, a short distance away. The third is that the market has stopped paying for the story. Ordell traded at A$1.06 within the past year. It closed at A$0.475 on 2 September 2026. Over the same window the gold price has been high by any recent standard, at about A$6,110 an ounce on 2 September 2026, although it is not at a record and Section 10 sets out why. A gold explorer whose share price has more than halved in an environment like that is either being correctly marked down or is being ignored. Both are worth understanding, and they lead to different conclusions. This workup sets out what Ordell has actually released, what that evidence supports, what it does not support, and what the maiden MRE has to show for the current story to hold. It is not a valuation and it does not contain a price target. Samso does not publish either. Figure 1 is the ground the whole argument rests on. Figure 1. Drilling at the Barimaia Gold Project, looking north-west. Source: Ordell Minerals investor presentation, 25 August 2026, cover slide. 2 The Numbers Ordell listed on the ASX on 19 July 2024 after raising $6 million. Two years on, the corporate position is simple and there is very little to untangle. Table 1. Ordell Minerals corporate snapshot Share price and market data as at the close on 2 September 2026. Balance sheet and share count as reported by the company at 30 June 2026, with the share count cross-checked against the August 2026 investor presentation. This is a living document and the share price will have moved by the time you read it. Three things in that table do the work. The 80.3 per cent. Barimaia is a joint venture and Ordell has earned 80.3 per cent of it, not all of it. Every ounce the maiden MRE reports will be a 100 per cent project figure, because that is how JORC resources are reported. Ordell's economic share of those ounces is four fifths. That is not a criticism, it is an arithmetic step readers should do themselves and one that is easy to skip when a headline resource lands. The cash. A$8.2 million at 30 June 2026 is a genuine balance, and it came from a A$7 million placement completed in the June quarter. The company's own Appendix 5B put estimated quarters of funding available at 5.0, calculated on the June quarter's outgoings of A$1.64 million. That calculation is now out of date in a specific and predictable way. Ordell has since started a 14,000m drilling program. Roughly 20,000m will be completed between May and September 2026. Exploration spending in the June quarter, when about 5,000m was drilled, was A$1.473 million. A quarter carrying two to three times that metreage will not cost A$1.473 million. Readers should assume the burn rate steps up sharply in the September and December quarters and should not carry the 5.0 quarters figure forward. The placement price. The May 2026 raising was done at A$0.63, a 7.4 per cent discount to the previous close of A$0.68 and a 16.5 per cent discount to the 15-day volume weighted average price of A$0.754. Institutional and sophisticated investors paid A$0.63 in May. The stock closed at A$0.475 on 2 September. Anyone who took that placement is currently underwater by about a quarter. That is a fact about market appetite, not a judgement about the asset, and it cuts both ways. It says professional money was willing to fund the drilling four months ago. It also says the market has re-rated the story downwards since. What the company is spending money on The June 2026 quarter is a fair sample. Exploration and evaluation absorbed A$1,473,000. Employee costs were A$98,000 and administration and corporate costs were A$106,000. Payments to related parties, being the Managing Director's remuneration and non-executive director fees, totalled A$105,000 for the quarter. That is a lean structure. Roughly 88 cents in every dollar of operating outflow went into the ground. For a company of this size that ratio is at the good end of what the ASX small end of town produces, and it is worth stating plainly because the opposite pattern is common enough to be worth checking for. One item deserves a note. Ordell's prospectus of 28 May 2024 allocated A$2,263,100 to exploration of Barimaia. Actual spending on Barimaia from admission to 30 June 2026 was A$5,708,000. Administration costs were budgeted at A$1,390,000 and actual spending was A$1,926,000. Goodia was budgeted A$1,136,000 and has absorbed A$360,000. The company has, in short, put far more into its flagship than it told the market at IPO it would, and correspondingly less into the second project. Investors can read that as focus or as drift depending on temperament. We read it as focus, because the extra money was funded by later raisings rather than taken from the original pool, and because the drilling results at Barimaia are the reason the money moved. 3 The Asset, and What Is Actually There Barimaia sits near Mount Magnet, about 550km north-east of Perth, in the Murchison district of Western Australia. It is a Tier 1 jurisdiction in the plain sense of the phrase. Title is secure, the roads exist, the skills exist, and there are several operating gold processing plants within trucking distance. Figure 2 shows how crowded that neighbourhood is, and Figure 3 shows how close Ordell sits to the largest operator in it. Figure 2. Barimaia in its district. The red star marks the project. Every labelled deposit within the 100km ring is an existing gold resource held by Ramelius Resources, Westgold or Glyden. Source: Ordell Minerals, June 2026 Quarterly Activities Report, 30 July 2026, Figure 1. The company holds two mining leases, an exploration licence and a spread of prospecting licences across the project, with three further prospecting licences added during the June 2026 quarter. Those three, referred to as the Wanbanna tenements, cost A$40,000 in cash and A$40,000 in Ordell shares, being about 58,737 shares issued at A$0.681. A deferred payment of A$250,000 becomes due to the vendor, Wanbanna Pty Ltd, if Ordell announces a JORC compliant Mineral Resource of more than 25,000 ounces of gold on those tenements. Ordell can settle that in cash or shares at its election. It is a small, sensibly structured deal, and the deferred trigger is set low enough that it will bind if the ground works at all. The geology, in plain terms The rocks at Barimaia are Archaean, which is the geological eon running from about 4.0 to 2.5 billion years ago. The greenstone belts of the Murchison were laid down within that eon, at about 2.7 to 3.0 billion years, and they belong to the Yilgarn Craton, the block of ancient crust that hosts most of Western Australia's gold. Within that block, gold in the Murchison is usually found in shear zones, which are bands of rock deformed and altered by movement, and in quartz lodes. Barimaia is being explored for a different target. The host rock Ordell is chasing is granodiorite, which crystallised from magma that pushed up into the older greenstone sequence and cooled at depth rather than erupting. Ordell has mapped and drilled a mineralised intrusive system with more than 7km of strike, which is the length of the target measured along its trend. The named prospects along that trend are McNabs, McNabs East, The Furnace and, on the newly acquired ground, untested extensions. Figure 3 puts that ground against Ramelius' holding. Figure 3. Barimaia against the Ramelius Mt Magnet ground. The orange outline is Ordell's project, the blue outlines are Ramelius tenements, and the black arrow measures 7.5km across the project, and the red dashed ellipse below it is the target zone through McNabs, McNabs East and The Furnace. Eridanus is labelled at upper left. Source: Ordell Minerals investor presentation, 25 August 2026, slide 4. The alteration at McNabs East is described by the company as sericite and pyrite dominant, with gold associated with bismuth, molybdenum, selenium, tellurium, tungsten, lead and silver. Mineralisation sits in quartz and pyrite extension veins with gentle north and south dips, and in biotite, carbonate and pyrite fracture veins dipping steeply south. Foliation dips moderately to the south. The company interprets the position of mineralisation as possibly influenced by steeply dipping, roughly east to west striking brittle to ductile faults. None of that description is unusual for an intrusion-related gold system and none of it is a proof of size. What it does establish is that Ordell has a coherent geological model rather than a scatter of hits. The elements associated with the gold, the vein sets and the alteration are consistent between prospects and between drilling programs. That matters when the maiden MRE is built, because a resource estimate is only as good as the geological model that controls the wireframes. 4 The Drilling Record, Tested Three drilling methods appear throughout this section and they are not interchangeable. Aircore is shallow and cheap. It samples the weathered top of the bedrock. It is used to find targets, not to define them. Reverse circulation, or RC, uses a hammer to produce rock chips and is the workhorse for defining a shallow deposit. Diamond drilling cuts a solid core of rock, costs the most, and is the only one of the three that lets a geologist measure structure, orientation and true width directly. Ordell's maiden resource will rest mainly on RC drilling, with diamond holes providing the structural control. Ordell has drilled about 55,000m since listing in July 2024. Roughly 20,000m of that will have been completed in the five months from May to September 2026. This is a company that drills. Michael Fowler, Ordell's Managing Director, set out the order of priorities himself when the program was announced. "Given our growing confidence in the continuity and robustness of the mineralisation at McNabs East within the Main Zone and Northern Zone, our immediate priority now is to complete the in-fill drilling required to deliver a maiden Mineral Resource Estimate in the December 2026 Quarter. This will, in turn, provide us with the foundation to assess the potential for a near-term development opportunity at Barimaia."Michael Fowler, Managing Director, Ordell Minerals, ASX release, "Major 14,000m Resource Drilling and Exploration Program Underway at Barimaia Gold Project, WA", 17 August 2026. Read that carefully and it is a statement about confidence in continuity, not about grade. The rest of this section tests whether the published drilling supports it. The results below are the ones the company itself puts forward as significant, taken from its own releases. We have grouped them by the announcement that carried them so that readers can see how the story built rather than seeing a single flattering list. Table 2. Selected drill intercepts at McNabs East and The Furnace Intercepts as reported by Ordell Minerals, grouped by the announcement that first carried them. Downhole intervals, not true widths. The company has not published true width estimates for most of these intercepts, so the actual thickness of mineralisation is unknown and will in most cases be less than the figure shown. "g/t Au" means grams of gold per tonne of rock. Reading that table honestly Two features of the McNabs East data set should shape how anyone reads it. The first is that the very high grades are concentrated in very short intervals. A 2m intercept at 92.20g/t gold containing 1m at 182.3g/t is a striking number. It is also a single metre of rock. Gold that behaves this way is described as nuggety, meaning it is distributed unevenly at a fine scale rather than being spread through the rock. Nuggety gold is harder to estimate, because a single sample can pull an average a long way. Resource estimators handle this by capping high grades, a process called top-cutting. Investors who anchor on the headline high-grade hits and then see a maiden resource reported at an average grade of 1.5 to 2.5 grams should not be surprised. The estimation process is designed to produce exactly that outcome. The second is that the intercepts are downhole lengths, not true widths. Ordell reports mineralisation dipping moderately to the south and holes drilled to test it. Without true width figures, the reported 22.7m and 29m intervals cannot be read as 22.7m and 29m of rock thickness. The company has not published a true width conversion for most of these results. This is normal industry practice at this stage and it is not a criticism. It is a caution, and it is the caution most often ignored by retail readers. Figure 4 is the plan view of what has been drilled so far. Figure 4. The McNabs East drill plan as at August 2026. Coloured dots are gram-metre intercepts, which is grade multiplied by width, so larger and redder dots carry more metal. The Main Zone, Northern Zone and The Furnace are labelled, along with the target zones between them. Source: Ordell Minerals ASX release, 17 August 2026, Figure 1. What the plan view shows, and what the numbers on their own do not, is that the mineralisation is not a single tight shoot. It is a broad corridor of intercepts running roughly east across more than 1km at McNabs East, with a separate cluster at The Furnace about 1km further east again, and a gap in between that Ordell is now drilling. The density of results in the Main Zone is high. The density at The Furnace is very low, because it has only had a first pass. Figure 5 is the section through the Main Zone. Figure 5. Long section through the McNabs East Main Zone at 6,891,900N, looking north. The pink shading is the interpreted mineralisation outline. The red arrows on the right mark where the system is open and untested. Source: Ordell Minerals ASX release, 17 August 2026, Figure 2. The section is the more useful picture. It shows near-surface mineralisation from roughly 10m below surface, which is what makes open pit mining a reasonable thing to test for. It also shows the system continuing to the east and to depth. One caution on reading it. This is a longitudinal section, meaning it is cut along the strike of the body rather than across it, so what it displays cleanly is plunge, the direction the ore shoot rakes within its own plane. Ordell's figure marks the same arrows as open down dip. Because the mineralisation dips moderately south, into the page, true down-dip continuity is only properly tested on cross sections cut at right angles to this one, and the company has published fewer of those. Hole 26BADD038 returned 11.0m at 1.06g/t gold from 227m to the end of sampling. That phrase means the assayed interval ended in mineralised rock, not that the hole did. The hole continued, and Ordell has stated that results from 238m to 281m are pending. The distinction matters, because a hole that ends in mineralisation and a hole whose assays simply have not come back yet are different pieces of evidence. "Open" is one of the most overused words in exploration. Here it is doing real work, and it also carries a cost. A system that is open at 230m depth is a system whose eastern extensions are unlikely to be mined by open pit. Depth extensions are good news for the total size of the system and largely irrelevant to the near-term development case that Ordell is building the maiden MRE to support. Those are two different stories and they should not be blended. 5 The Science Tested The Samso research method requires that the geological claim underneath an exploration story be checked against the published science rather than accepted from the company. Ordell's claim, reduced to a sentence, is that gold in the Murchison can sit in granitoid intrusions and that Barimaia's granodiorite is therefore a legitimate target. That claim has a peer-reviewed foundation and it is older than the company. "In the Archaean Murchison Province of Western Australia, granitoid batholiths and plutons that intruded into the ca. 2.7–2.8 Ga and ca. 3.0 Ga greenstone belts can be divided into three major suites... gold mineralisation in the Murchison Province was temporally and spatially associated with Suite II I-type granitoids in the northeastern part of the province."Wang, L.G., McNaughton, N.J. and Groves, D.I., 1993. "An overview of the relationship between granitoid intrusions and gold mineralisation in the Archaean Murchison Province, Western Australia", Mineralium Deposita, volume 28, pages 482 to 494. So the model is real. Gold and granitoids in the Murchison have a documented spatial and temporal relationship in the academic literature going back more than thirty years. Two qualifications are needed, and we would rather state them than let the citation do more work than it can carry. The first is that Wang and co-authors tie the gold to Suite II, which they classify as I-type trondhjemite and tonalite bodies. Ramelius describes the Eridanus host as a granodiorite. Whether the Eridanus Granodiorite maps onto Wang's Suite II or onto their Suite I monzogranite and granodiorite suite is not something we could establish from published sources. We are not asserting the correspondence, and we would go a step further. On a strict reading, the suite the paper favours for gold is not the rock type Ramelius names at Eridanus. So the literature supports the general model of intrusion-associated gold in this province. It does not specifically endorse granodiorite as the favoured host. The qualification that cuts the other way is that field names for intrusive rocks are approximate. A body logged as granodiorite in the field can classify as a tonalite or trondhjemite once it is properly analysed, and nobody has published that analysis for either deposit. The second is more useful. The strongest evidence for the model is not the paper. It is Eridanus itself. Ramelius describes it in its own words. "Mineralisation at Eridanus occurs as stockwork veins concentrated within the east-west orientated Eridanus Granodiorite, which has intruded into felsic aphyric to porphyritic intrusive rocks."Ramelius Resources, Eridanus project page, rameliusresources.com.au, accessed 2 September 2026. Ramelius also notes that this bulk-tonnage style distinguishes Eridanus from the higher-grade lode deposits found elsewhere in the Mt Magnet field. That is the point. A 1.3 million ounce granodiorite-hosted deposit is in production 6km away. The model is not theoretical in this district. It has been mined. What this does and does not establish is worth being precise about. It establishes that the target type exists at economic scale in this specific belt. It does not establish that Barimaia contains one. Proximity to a deposit and sharing a host rock type are the reasons to drill, not evidence of what the drilling will find. The intercepts in Section 4 are the evidence, and they are sufficient to justify a resource estimate. They are not yet sufficient to establish scale. 6 Metallurgy and the Development Path Metallurgy is where a lot of gold stories quietly fail, and it is usually tested far too late. Ordell tested it early, and the results are the strongest single piece of technical evidence the company has released. Table 3. McNabs East metallurgical testwork results Cyanide leach testwork on representative samples of transition and fresh rock from McNabs East, completed by independent laboratory ALS Metallurgy Pty Ltd. "P80" is the grind size at which 80 per cent of the material passes the stated screen, measured in microns. A finer grind costs more power. "Extraction" is the percentage of contained gold recovered into solution after the stated leach time. Four things in that table matter and none of them are marketing. Extraction above 95 per cent after 48 hours at a relatively coarse 106 micron grind means the gold does not need to be ground fine to be recovered. At 24 hours the fresh sample sits just under that mark, at 94.6 per cent, so the 95 per cent figure is a 48-hour number and should be quoted as one. Comminution, meaning crushing and grinding, is typically the largest single power consumer in a gold plant, so a coarser grind is a direct saving on operating cost. Gravity recoverable gold of 34.9 to 40.8 per cent across the four tests, a range Ordell itself states as 35 to 40 per cent, means a third to two fifths of the metal can be caught by density alone, before any chemistry, which reduces reagent consumption. Laboratory gravity recoverable gold is a theoretical maximum and a plant will realise less of it, so treat that range as an upper bound rather than a design figure. Cyanide and lime consumption came in below 0.4kg per tonne, which is low. And the Bond ball mill work index, a standard hardness measure, was 10.8kWh per tonne for transition material and 13.8kWh per tonne for fresh, both in the moderate range. Taken together the description that fits is "free milling", meaning the gold is not locked inside sulphide minerals or otherwise refractory. Free milling ore can go through a conventional crush, grind and carbon-in-leach circuit. There are hundreds of those in Western Australia, including several within trucking distance of Barimaia. There are two honest caveats and the second is the more important one. The first is sample size. This is four leach tests on two rock types from one prospect, on head grades of 1.03 and 1.63g/t. It is early-stage testwork done properly, not a metallurgical program supporting a feasibility study. The second is that the samples were transition and fresh material only. No oxide was tested. Oxide is the fully weathered rock at the top of the profile, and in an open pit it is the first material mined. Ordell's own shallow intercepts start about 10m below surface, which puts a meaningful part of the near-surface mineralisation in exactly the zone that has not been through a leach test. Oxide ore usually leaches well and often better than fresh rock, so this is a gap rather than a warning. It is still a gap, and it is the obvious next piece of testwork. What the testwork does is remove a common way for a project like this to die. It does not, on its own, prove anything about the size or the economics of a future operation. Why the metallurgy matters more here than usual Ordell's stated strategy has two prongs. One is to fast-track a potential near-term development opportunity underpinned by the maiden MRE. The other is to keep growing the system through extensional and discovery drilling. The near-term development prong is only credible for a company with A$8.2 million in the bank if the ore can be treated in someone else's plant. For a sense of the capital involved, Ramelius is spending about A$223 million to expand a plant it already owns. Building one from nothing is not on the table for a company of Ordell's size. Trucking ore to an existing plant is. Here we need to be careful, because this is the point where commentary on companies like Ordell usually gets ahead of the facts. Ramelius has never publicly said it will buy or toll treat third-party ore from the Murchison. We looked. We did not find such a statement, and readers should not assume one exists. What is documented is that Ramelius is spending about A$223 million expanding the Mt Magnet processing plant from 1.9Mtpa to 4.3Mtpa, with the two circuits targeted for operational readiness by September 2027. What is also documented is that the practice exists in the district. Westgold has an arrangement to process ore containing roughly 140,000 ounces over two and a half years for New Murchison Gold through its Bluebird plant. So the defensible statement is this. There is expanding mill capacity in the district, there is an established precedent for third-party ore processing in the Murchison, and gold prices are high enough that trucking ore over meaningful distances can work. There is no agreement, no discussion and no commitment involving Ordell that has been made public. Anyone building a case on a Ramelius ore purchase is building it on inference. 7 The Eridanus Analogue, and Where It Stops The comparison to Eridanus is the centre of the Ordell story and it is used, correctly, by the company itself. Michael Fowler has framed the strategy as trying to "find an Eridanus" by systematically drilling over 2.5km of strike. Set out plainly, here is what the two have in common and where the comparison ends. Table 4. Barimaia and Eridanus compared A like-for-like comparison of the two deposits on the points that matter to an investor. Eridanus figures are Ramelius Resources' reported Mineral Resource. Barimaia has no reported Mineral Resource, so the corresponding cells state what is and is not known rather than a figure. The left column is a mine. The right column is a drilling program. Everything in the middle of that table is the work Ordell has yet to do, and the maiden MRE is the first of those steps, not the last. We would put the analogue this way. Eridanus tells you the target type is real and economic in this belt. It does not tell you Barimaia is one, it does not tell you how big Barimaia will be, and it does not tell you whether Barimaia's grade will support the same open pit economics. Companies use analogues because they compress a complicated geological argument into one memorable comparison. That compression is legitimate as a reason to explore. It becomes misleading the moment it is treated as a forecast. 8 Comparables Ordell has no resource, which makes conventional peer comparison awkward. The honest approach is to show where similar companies sit and let readers do the arithmetic themselves. Samso publishes no valuations and no price targets, and we are not going to derive an implied value for Ordell from this table. The table is context, not a conclusion. Table 5. Ordell and six comparable ASX-listed gold companies A set chosen for comparable stage and geography rather than comparable size. Every market capitalisation comes from a single reading of the ASX company data feed at 06:06 AWST on 4 September 2026, reflecting the close on 3 September 2026. That is one day later than the 2 September close the rest of this workup is built on, so Ordell appears here at A$0.495 rather than the A$0.475 used in the snapshot. Resource figures are each company's most recent published JORC estimate, with the date given. Two of these companies are no longer straightforward comparables and are marked as such. Figure 6 draws the same table, because the shape of it is the argument. Figure 6. The same set as Table 5, drawn. Ordell is the orange bar. The italic column on the right is each company's reported JORC resource, which is what the market is being asked to pay for. Ordell is one of three companies shown with no resource at all. All market capitalisations are from one reading on 4 September 2026, so unlike the earlier version of this chart the comparison is same-day. Samso Research Media House, built from the sources in Table 5. Three observations, offered as observations rather than as a case. Ordell sits near the bottom of this group on market capitalisation and at the lowest end on resource, because it has none. That is internally consistent. A company without a resource trading below companies with one is not an anomaly requiring explanation. The genuine like-for-like comparison is narrower than it first appears. Of the six other companies, Santa Fe Minerals is no longer a Western Australian gold story at all. It bought the Eburnea project in Côte d'Ivoire from Turaco Gold in July 2025, and its March and June 2026 quarterly reports record that work at Challa in the Murchison was limited to desktop studies in both periods. The page title on its own announcements page reads "ASX Listed Shell". Caprice Resources is the closest true comparable, in the Murchison, with no resource, drilling hard, and holding granted mining leases. Ordell and Caprice are the two companies here being asked the same question at the same time, and Caprice carries roughly twice Ordell's market capitalisation. The spread between Great Boulder at roughly A$191 million and Torque Metals at roughly A$134 million is a caution against simple ounce counting, and it is more instructive than it first looks. Great Boulder holds 1.5 million ounces across two Western Australian projects. Torque holds 351,000 ounces in Western Australia and a further 1.5 million ounces at Edleston in Ontario, at about 1.0g/t. So the gap in market capitalisation is not simply an ounce gap, although the two sets of ounces are not equivalent, and Torque's larger parcel is lower grade and on another continent. Grade, depth, metallurgy, ownership percentage, tenure and permitting all move the number, and an ounce in one deposit is not an ounce in another. We flag this because the temptation, once Ordell publishes a resource, will be to divide market capitalisation by ounces and reach a conclusion. That arithmetic is easy and frequently wrong. One caution on this whole section. We could not find a credible published benchmark study of dollars per resource ounce for Western Australian gold juniors current to 2026. Any multiple a reader constructs from the table above is their own calculation, not an industry standard, and it should be treated that way. 9 Where Ordell Sits, on Five Measures Table 5 compares what these companies own. It does not compare what stands between them and a mine. This section does that, across five measures, and it is built from each company's own filings rather than from our impression of them. The fifth measure is a judgement and is treated as one. A note on what this section is not. Ranking seven companies by which is most likely to make an economic discovery would be a valuation dressed as geology, and Samso does not publish those. What follows sets out the evidence each company has produced and what each one still has to prove. Readers who want a ranking will have to build it themselves, which is the point. Table 6. Five measures across the comparable set Market capitalisations as in Table 5. Stage is the furthest point each company has formally reached, not what it hopes to reach. Jurisdiction risk covers tenure, native title and permitting as each company has disclosed them, rather than country risk, which for six of the seven is the same. What the first four measures show Two of these companies are not really explorers. Saturn Metals has an Ore Reserve, a completed pre-feasibility study and a definitive study running, and Western Gold Resources has taken a formal Decision to Mine with its mining proposals approved. Comparing Ordell to either on market capitalisation compares a company that has not yet counted its metal with companies that have costed extracting it. Two more are ahead of Ordell on resource but not on permitting. Great Boulder has 1.5 million ounces across two projects and a market capitalisation five times Ordell's, but all three of its Side Well mining leases are still applications, its native title agreement is still under negotiation, and it has no binding arrangement to process Side Well ore. Torque Metals has 351,000 ounces in Western Australia and, on its own description, no study of any kind. Ounces in the ground and a path to mining them are different assets, and the market does not always price them differently. That leaves Caprice Resources as the one company here being asked the same question as Ordell at the same time, and Santa Fe Minerals as a company that has left the comparison without the table noticing. On commodity the set is uniform, and that is the point of it. Six of the seven are gold companies in Western Australia, which is what makes the comparison worth making at all. Where a company carries something else, it is minor and the company treats it as such. On jurisdiction the differences are real but they are not country differences. Every one of these projects except Santa Fe's active ground sits in Western Australia, which is about as settled a mining jurisdiction as exists. The differences are at the tenement. Ordell owns 80.3 per cent of its flagship, so a fifth of anything it finds belongs to someone else. Great Boulder is in the same position and it is worth saying so, because most of the Side Well deposits are 75 per cent owned under one joint venture agreement and Golden Bracelet 80 per cent under another, while the resource is quoted on a 100 per cent basis. Western Gold does not hold its tenements at all. They belong to GWR Group, and Western Gold works them under a deed of co-operation with a royalty back to the holder. Great Boulder's leases are applications. Caprice and Ordell both hold granted mining leases, which is the cleanest position in the table. Figure 7. The first four measures, drawn. A filled circle means the company has formally reached that stage and said so in its own filings. Ordell, Caprice and Santa Fe have reached only the first. Nobody in this set has reached all five. Western Gold holds mining approvals and a Decision to Mine without ever having declared an Ore Reserve, and Saturn has the Ore Reserve but not the approvals, so both of those rows have a gap in them. Samso Research Media House, built from the sources in Table 5 and Table 6. The fifth measure, geological potential for an economic discovery This is where the comparison stops being arithmetic. What follows is each company's geological case as its own drilling supports it, and the specific thing that would have to be true for that case to become a mine. Ordell. The strongest analogue in the set. Ramelius Resources' Eridanus pit sits about 6km to the north-west, in granodiorite, with a Mineral Resource of 24Mt at 1.7g/t for 1.3 million ounces. Ordell is drilling the same host rock along more than 7km of strike, and its metallurgy returned better than 95 per cent extraction after 48 hours at a 106 micron grind, with 35 to 40 per cent recoverable by gravity alone. What has to be true is continuity at close spacing, and that is exactly what the current infill programme tests. The unresolved problem for Ordell, set out in Section 14, is that the infill grid stops at about 60m to 90m vertical while the celebrated deep intercepts sit at 137m to 227m downhole in angled holes, so the separation is indicative rather than exact. Either way the higher-confidence part of the maiden estimate will be shallow and comparatively low grade. Great Boulder Resources. The largest reported endowment in the set after Saturn, and a high-grade one at 2.0g/t. The geology is well described, with two intersecting lode sets at Mulga Bill where structural blowouts produce the thick high-grade intervals, and the deposit is open in all directions with a step-out hole 430m below surface returning 1.93m at 574g/t. The constraint is not geological. Thirty-nine per cent of the ounces are Inferred, none is Measured, the only economic study covers 79,000 ounces of the million, and the metallurgy is proven on one deposit out of six. Torque Metals. Narrow orogenic lodes of 1m to 10m, averaging 2m to 5m, at the project Torque renamed Ritz in August 2026, in the Kalgoorlie Terrane beside the Boulder-Lefroy Shear Zone, with free-milling metallurgy at approximately 96 per cent. The company discloses a nugget effect of 20 to 40 per cent, and two holes at the HHH deposit show what that means in practice. One reverse circulation hole returned 11m at 456g/t. A diamond hole collared about 3m away returned 7.9m at 1.37g/t. Grade continuity, not grade, is the question here, and the resource grade has already fallen from 4.3g/t to 3.8g/t as drilling tightened the model. Saturn Metals. The most advanced case and the one that depends least on further discovery. Apollo Hill is 2.83 million ounces at 0.51g/t, a bulk-tonnage deposit whose entire economics rest on cheap mining at a 2.39 to 1 strip ratio feeding a heap leach. Column tests averaged about 78 per cent at an 8mm crush, and the pre-feasibility study carries 73.7 per cent over a 160-day leach at operational scale. At 0.47g/t reserve grade there is very little margin for a recovery shortfall, and heap leaching is, in the company's own words, less frequently used in Western Australia, which it flags as a schedule risk across a long list of approvals still to come. This is an engineering and approvals question now, not an exploration one. Caprice Resources. The nearest thing to Ordell in the table, and the higher-grade headline result of the two. Vadrians runs more than 1,000m of strike to at least 400m depth and is open, hosted in banded iron formation, with one intercept first reported as 22m at 66.2g/t from 42m on two-metre composites and since restated, on one-metre splits, as 19m at 124.7g/t from 42m. The company estimates true widths at 60 to 70 per cent of downhole length and says plainly that the host lithology of its best hole is not yet clearly resolved. It has no resource, its Exploration Target predates the discovery that matters, and it is drilling scissor holes to work out the orientation of its own best intercept. Santa Fe Minerals. On the Western Australian ground there is little to assess, because almost no work is being done. Auger sampling returned peak values of 256 parts per billion gold, which is anomalism rather than mineralisation, and the best historic drill result the company reports at Golden Girls is 3.74g/t. Its stated next step is target-specific aircore drilling, and the last two quarterly reports record desktop studies instead. The active exploration is in Côte d'Ivoire, where the best historical intercept is 26m at 4.82g/t at Satama. Of its Bouake North ground the company says drilling is not yet sufficient to establish continuity of individual mineralised zones, which is a fair description of the whole portfolio. Western Gold Resources. The most complete small deposit and the most constrained. Gold Duke is 277,000 ounces, entirely oxide, at 1.8g/t, in narrow banded iron formation lodes averaging 2m to 2.4m at Eagle. The Competent Person's own words are that the mineralisation is nuggety, exhibits short continuity ranges, and that this short-range variability and the uncertainty that goes with it will be defining factors for the Gold Duke deposits. Grade continuity is not known to exceed 60m. The resource is capped by an economic pit shell rather than by geology, with published pit designs reaching between about 27m and 60m, nothing below the base of oxidation has been metallurgically tested, and the grade fell from 2.1g/t to 1.8g/t when the company drilled it more densely. Denser drilling lowering the grade is the risk every pre-resource company in this table carries, and Western Gold is the one that has already demonstrated it. The one comparison worth carrying forward. Two companies in this table have published a lower grade after more work. Western Gold's Gold Duke estimate went from 2.1g/t to 1.8g/t over a year in which it drilled 33,980m on 5m by 10m spacing, although the company attributes the fall to an updated geological interpretation, revised classification and the inclusion of additional material rather than to the drilling density itself. Torque's Paris deposit fell from 4.3g/t to 3.8g/t when tighter domaining was applied. Neither is a failure, and neither is evidence about Ordell. What they show is that a grade published from wide-spaced holes is a working number, and that the direction of travel when a deposit is examined more closely is often downward. That is worth holding in mind between now and the December quarter, and it is not a prediction. 10 The Market Behind the Metal Gold is the whole macro case here, so it deserves accuracy rather than enthusiasm. As at 2 September 2026 gold traded at about US$4,388 an ounce, up roughly 23 per cent over twelve months. In Australian dollars the price was about A$6,110 an ounce, with Ordell's own website quoting A$6,125 the same day. The two sources agree closely enough to use either. Now the correction that most commentary written this week will get wrong. The Australian dollar gold price is not at a record. It peaked in January 2026, fell to a low around A$5,680 an ounce in mid-July 2026, and has recovered since. The current price sits roughly a fifth below that January high. We have seen the January peak quoted at two materially different levels by two different sources and we are not going to print a figure we cannot stand behind. The direction and the timing are reliable. The peak number is not. The twelve-month gain in Australian dollars, at about 12 per cent, is also smaller than the US dollar gain of about 23 per cent, because the Australian dollar has appreciated over the period. Australian gold producers earn in Australian dollars, so that gap matters to them and, eventually, to anyone hoping to sell them a deposit. What this means for Ordell is straightforward. The gold price is high in absolute terms and has been for long enough to change corporate behaviour. Mills are being expanded. Ore is being trucked further than it used to be. Producers are looking for feed. That environment is favourable for a company with an undeveloped deposit next to a growing plant. It is also an environment that is already priced. A A$6,100 gold price is not news, and it did not stop Ordell falling from A$1.06 to A$0.475. Investors who expect a strong gold price to carry an explorer's share price on its own have the last eight months of Ordell's chart as evidence against them. 11 Goodia and Fisher South Ordell owns two other projects outright. Both are early. Neither is the reason to look at the company today, and we would be doing readers a disservice by inflating them. Goodia Goodia sits about 15km west of Norseman, in a different part of the Yilgarn from Barimaia, with more than 30km of prospective greenstone. Ordell has completed soil geochemical sampling over 10km of strike, collecting 909 samples in the May 2026 program alone at line spacings of 200m, 400m and 800m, with sample points 50m to 100m apart along each line. Thirteen rock chip samples were also collected. The sampling has defined extensive surface gold anomalism along the eastern margin of the Goodia Dome, with anomalous zones up to 1.8km long and open ended. These sit south of three earlier anomalies, GD1 to GD3, reported in November 2025. First-pass aircore drilling is planned to test them. A soil anomaly is a pattern of slightly elevated gold in surface material. It is a reason to drill and nothing more. Ordell has not drilled Goodia. Until it does, the honest description of Goodia is a well-defined set of untested targets on ground the company owns entirely. Fisher South Fisher South is 400km north of Kalgoorlie in the Laverton district, about 40km along strike from Emerald Resources' Dingo Range project. Ordell completed a 59-hole aircore program for 2,717m on a very wide 1,600m by 200m grid in the June 2026 quarter. Holes averaged 46m deep and the objective was to understand stratigraphy and regolith under transported cover, not to find a deposit. The best gold result was 14m at 150 parts per billion gold from 45m in hole 26FSAC021, at the northern end of a gold anomaly defined by an earlier explorer's auger drilling in 2011 that extends over 1.6km to the south. To be clear about scale, 150 parts per billion is 0.15 grams per tonne, which is roughly a tenth of the grade at Eridanus. It is an anomaly, not an intercept. The more interesting result may be the tungsten. Bottom-of-hole samples on the western side of the drilled area returned 1m at 1,718 parts per million tungsten in hole 26FSAC059 and 1m at 200 parts per million in 26FSAC053. Ordell states that further work is required to understand the source and significance of this. We agree, and we would add that a single strongly anomalous tungsten sample in a first-pass regolith program is a long way from anything, however unusual the number looks. A note on the source documents. Ordell's 2 July 2026 announcement contains two small internal inconsistencies worth recording. The text describes 13 rock chip samples while the accompanying table lists 15 sample identifiers, one of them unassayed. The text gives the aircore hole depth range as 24m to 84m while the collar table lists one hole at 20m. We report both as published. Reconciling them is a matter for the company, and neither affects the substance of the result. 12 Management Weighed The single most striking thing about Ordell's board is that it is largely the Genesis Minerals team, reassembled. Table 7. Ordell Minerals board and management Roles and the verifiable career history behind each. Shareholdings are those disclosed in the FY2025 Annual Report and will have changed. Where a claim commonly made about a director could not be verified from a primary source, it is left out rather than repeated. What the record supports Fowler's Genesis history is the most directly relevant item on the board. Genesis acquired Barimaia in 2015, on the basis that its geology resembled the Ulysses deposit, and later let it become a lower priority. Fowler ran Genesis at the time. When Ordell was formed, it bought Barimaia from Genesis, and Genesis took shares as part of the consideration. The man now drilling the project is the man who bought it for Genesis in the first place. On his own account the reason he came back to it is straightforward. He has described Barimaia as "highly prospective" and said it "warrants a lot more work". That is a stated view rather than a demonstrated one, but it is at least his own, on the record, about an asset he has known for a decade. That is a genuine signal and we would rate it highly. It is not the same as a track record of discovery. Growing a resource from 0.12Moz to more than 2Moz at Ulysses is a real achievement in resource growth. It is a different skill from finding a deposit where none was known, and readers should hold the distinction. McKeith's presence is the corporate signal. A current Lead Independent Director of Evolution Mining and a newly appointed director of Perseus does not need to chair a A$36 million explorer. He has also chaired Genesis. The pattern across this board is people who have already done the small-company-to-real-company transition once, together. What we are not saying Board quality is the most over-weighted variable in small resources analysis. A good board improves the odds of a company handling a discovery well. It does not put gold in the ground. The Ordell board is better than average for a company of this size, and if the drilling does not deliver, that will not save it. We also want to record what we could not verify. It is frequently written that Tommy McKeith led teams to several significant gold discoveries. We could not find a primary source naming any of them. We are not repeating the claim as fact. Michael Fowler's Competent Person status is confirmed by the JORC statements in Ordell's own announcements, where he is named as a Member of the AusIMM, a Director and a shareholder. That combination is disclosed and compliant, and it is also worth readers noticing. The person signing off on the exploration results is an executive director and a substantial shareholder in the company reporting them. This is entirely normal in junior exploration and entirely legal. It is still a structure where independence rests on the individual rather than on the arrangement. 13 The Register, the Escrow and the Share Price Ordell's share register is tight and that is a double-edged feature. Table 8. Substantial holders and capital structure Share counts are as disclosed by Ordell Minerals at 30 June 2026. The substantial holdings are taken from the holders' own Form 604 notices lodged with the ASX, which we have read, rather than from the company's summary or from a data aggregator. The dates of those notices are given in the table, because two of the three are current and one is not. A 75.5 million share register is small. That is what makes Ordell a leveraged exposure to its own drill results, and it is what the company means when it calls itself tightly held and highly leveraged. The same feature makes the stock thin. A small register with limited institutional depth moves a long way on modest volume in both directions, and the 52-week range of A$0.345 to A$1.060 is that fact drawn as a chart. Two items on the register need to be read carefully. The Genesis holding is not a strategic endorsement. Genesis Minerals came onto the register because it sold Barimaia to Ordell and took shares as part of the price. Ordell's FY2025 Annual Report records 6,800,000 shares valued at $1,360,000 and 2,000,000 options issued to acquire the project from Genesis during the IPO process. So the position began as vendor scrip. What has happened to it since is more interesting than that, and it needed the notices to see. On 18 March 2026 Genesis exercised 2,000,000 options at 30 cents, paying $600,000 in cash, which took it from 4,000,000 shares and 6.5 per cent to 6,000,000 shares and 9.37 per cent. On 22 May 2026 it did not take part in Ordell's placement, and the same 6,000,000 shares were diluted back to 7.96 per cent. Read together, that is a holder taking up options it already had rather than one building a position, and then declining to follow the next raising. It is a decision, not passive legacy scrip, and it is also not accumulation. Genesis operates at Leonora and Laverton and has no Murchison asset. Reading the Genesis holding as evidence that a producer is positioning itself around Barimaia is a misreading. If a producer were to have a natural geographic interest in this ground, it would be Ramelius, whose pit and mill are next door, and Ramelius does not appear on the register. The Lowell figure is out of date and there is no newer one. Cremorne Capital, as responsible entity for the Lowell Resources Fund, last lodged a notice on 31 July 2025 showing 4,692,393 shares and 7.62 per cent. It has lodged nothing since. Between that date and now, Genesis exercised options in March and Ordell issued shares in the May placement, both of which enlarged the register. Unless Lowell has bought, its percentage today is lower than 7.62 per cent. No notice states what it is, and we are not going to calculate a figure and present it as a holding. The escrow has rolled off. Six million shares came out of 24-month IPO escrow on 19 July 2026, and Ordell lodged a release of securities from escrow notice on 9 July 2026. That is about 8 per cent of the register becoming free to trade in a stock with limited liquidity. Whether it has been sold, we do not know. It is a supply factor sitting behind a share price that has drifted since July, and it is worth naming rather than reaching for a story about sentiment. The price itself Ordell peaked near A$1.00 in late January and early February 2026, on the back of the January announcement of shallow, high-grade gold at McNabs East including what the company described as a bonanza grade hit. It has fallen through the year since, with a secondary peak in May around the placement, and has spent recent months in a band roughly between A$0.45 and A$0.55. Three explanations are available and they are not mutually exclusive. The stock ran hard on a single very high-grade intercept and gave that back as subsequent results proved good but not exceptional. The Australian dollar gold price fell about a fifth from its own January peak, taking the sector with it. And the escrow release added stock into a thin market in July. What we would not conclude is that the market has passed judgement on the maiden MRE. There is no resource to judge yet. What the price does say is that the market is not currently paying in advance for one, which is a different observation and, for anyone forming a view before the resource lands, a more useful one. 14 What Has To Go Right, and What Can Go Wrong What has to go right The maiden MRE has to arrive in the December 2026 quarter and it has to be big enough and shallow enough to support a mining study. The drill spacing tells you what kind of resource it will be. In the Main Zone, Ordell is tightening to 25m between sections and 20m between holes along each section, over 350m of strike, to about 90m vertical. In the Northern Zone the spacing is wider at 40m between sections and 40m or 20m along them, over 835m of strike, to about 60m vertical. The JORC Code prescribes no spacing for any confidence category, and classification is the Competent Person's judgement rather than a rule. As a general matter, though, the tighter Main Zone grid is the sort of spacing that can support Indicated material in a deposit of this style, and the wider Northern Zone grid is more likely to support Inferred. Only Indicated and Measured material can be converted into an Ore Reserve, so the Main Zone is where a mining study would have to be built. Figure 9 shows where those holes are going. The point most readers will miss. That infill drilling stops at about 60m to 90m below surface. The deep high-grade intercepts, including 4.6m at 19.60g/t from 177m and 13.3m at 2.37g/t from 195.1m, sit two to three times deeper than the grid that is being drilled to support the estimate. They are extensional and discovery results, not maiden resource feed. Whatever the maiden MRE reports, its higher-confidence material will be shallow and comparatively low grade, and the deep high-grade zones will either sit in a low-confidence category or fall outside the estimate entirely. Anyone anchoring on the 19.60g/t number and expecting the resource grade to reflect it has misread which part of the deposit is being counted. Figure 8 shows the separation. Figure 8. The same point as a picture. The shaded boxes are the depth the resource infill drilling actually reaches. The orange marks are the deep intercepts Ordell has reported, including the 4.6m at 19.60g/t that is quoted more than any other number in this story. They sit two to three times deeper than the grid being drilled to support the maiden estimate. Note the caveat printed on the figure. Infill depths are vertical, intercept depths are downhole along angled holes, so the separation shown is indicative rather than exact. Samso Research Media House, built from Ordell Minerals ASX releases of 29 April, 30 July and 17 August 2026. Figure 9. The infill drilling planned for August and September 2026. Blue squares are proposed reverse circulation holes. The tighter the spacing, the higher the confidence class a resource estimator can assign. Source: Ordell Minerals ASX release, 17 August 2026, Figure 6. Grade continuity has to hold at that tighter spacing. This is the real test of the program and it is the one investors most often skip past. Wide-spaced drilling that finds high grades and infill drilling that confirms them are two different results, and the second is much harder to achieve. Ordell's own stated objective for the Main Zone drilling is "to confirm continuity of mineralisation, particularly higher-grade zones identified to date". That is the company naming the risk itself, which we would rather see than not. And a processing route has to emerge. A resource without a plant is an asset held for someone else's benefit until a deal is done. What can go wrong The resource comes in smaller or lower grade than the drill headlines imply. This is the most likely bad outcome. It would be a re-rating rather than a failure. Nuggety high-grade intercepts get top-cut. Downhole widths become true widths. Ounces in a resource are always fewer than the sum of the excitement that preceded them. Two words in that paragraph do a lot of work, so here is what they mean. Top-cut, also called grade capping. Gold is unevenly distributed, and a deposit like McNabs East throws up the occasional extreme assay. Ordell has reported 1m at 182.3g/t and 0.8m at 89.43g/t. When a geologist builds a resource model, each drill sample is used to estimate the grade of the rock around it. If a single 182g/t sample were allowed to do that unchecked, it would push a very high grade across a large volume of rock that almost certainly does not contain it. So the estimator replaces the extreme values with a lower ceiling before estimating. That is the top-cut. The consequence for a reader is simple. The average grade in the reported resource will be lower than the average of the raw drill results, sometimes materially lower, and that is not the company being conservative. It is standard practice and a resource that did not do it would be the one to worry about. True width. A drill hole almost never meets a mineralised zone at right angles. It cuts through at an angle, so the length of the intersection measured down the hole is longer than the actual thickness of the zone. That real thickness, measured square to the body, is the true width. A 22.7m intercept reported down the hole could be a body considerably narrower than 22.7m, and how much narrower depends on the angle, which depends on the geometry. Every intercept quoted in this workup is a downhole length, because that is what Ordell has published. True widths are not yet known, and they will be, because a resource model has to work in real thickness rather than in drill metres. Again this is normal reporting, not a company failing. It is a reason not to picture the deposit as thick as the intercepts sound. The company has to raise money on a soft share price. Ordell had A$8.2 million at 30 June and is spending it fast on a 14,000m program. Its own Appendix 5B estimate of 5.0 quarters of funding was calculated on a quarter with a third of the current drilling rate. A raise during 2027 should be treated as the base case, not the downside case. Whether it is done at A$0.475 or at a post-resource price is the question, and the answer is not in the company's control. The 80.3 per cent gets forgotten. Ordell's share of any Barimaia resource is four fifths. The remaining interest sits with the joint venture partner. The open pit turns out to be smaller than it sounds, or does not stand up at all. Ordell has drilled shallow mineralisation from about 10m below surface and describes it as having potential to be extracted by open pit. Nobody has yet tested that with a pit optimisation, which is the engineering exercise that works out how much waste rock has to be moved to reach the ore and whether the result makes money at a given gold price. No pit optimisation has been published, no strip ratio has been published, and no mining study has been completed. The word "potential" is doing all the work in that sentence. Gold falls. The Australian dollar gold price has already fallen about a fifth from its January 2026 peak. There is no reason to assume it has stopped. The other projects absorb attention. Goodia and Fisher South are 100 per cent owned, which makes them tempting. Barimaia is 80.3 per cent owned and is the only asset near a resource. Spending on the two early projects while the flagship needs infill drilling would be a strategic error. There is no evidence Ordell is making it. The prospectus comparison in Section 2 shows the opposite pattern, and Figure 10 is what that spending buys on the ground. Figure 10. Reverse circulation drilling at the McNabs East Main Zone in May 2026, with the town of Mount Magnet on the horizon. The proximity in this photograph is the practical argument for Barimaia. Source: Ordell Minerals ASX release, "Ordell Expands Prospective Footprint at Barimaia Gold Project, WA", 20 May 2026, Figure 3. 15 What Would Change Our View We would rather commit in advance to what we are watching than explain afterwards why the outcome fitted the thesis. Things that would strengthen the case. A maiden Mineral Resource delivered in the December 2026 quarter as promised. A substantial share of that resource classified as Indicated rather than Inferred. An average grade at or above about 1.5g/t across a shallow, contiguous body. Infill results that confirm the higher-grade zones rather than diluting them. Any announced arrangement, at any stage, with an owner of processing capacity. Assays from the pending holes, including 26BADD038 below 238m, that extend the system without moving it deeper than a pit can reach. Things that would weaken it. The MRE slipping out of the December quarter without a clear reason. A resource dominated by Inferred material. Infill drilling that fails to reproduce the grades from wide-spaced holes. A capital raising at a discount to the current price before the resource is published. A pit optimisation, when one appears, showing a strip ratio that makes the shallow material uneconomic. Any material change to the 80.3 per cent joint venture interest. Things we would ignore. Single very high-grade intercepts. Soil anomalies at Goodia. Tungsten at Fisher South. Gold price commentary. None of these will decide whether Ordell works. 16 Samso Concluding Comments Ordell Minerals is an exploration company doing the ordinary work of exploration well, at a moment when the market has stopped paying attention to it. The case for looking now is a timing case rather than a value case. Ordell has spent two years drilling and has put the first formal count of what it has found on the calendar for the December 2026 quarter. Between now and then the company will publish infill assays, and after that a resource. That is a defined sequence of events with a defined end point, on a stock that has already given back most of what it gained in January. Investors who wait for the resource will be reading the same news as everybody else on the day it lands. Investors who look now are looking at a company where the evidence is already public and the conclusion is not yet drawn. What supports that timing. A deposit of the same style, in the same host rock, is being mined 6km away, which is as close to a working analogue as an explorer gets. The metallurgy was tested early and came back free milling, with extraction above 95 per cent after 48 hours at a relatively coarse grind, which removes one of the standard ways a project like this dies. The drilling is systematic rather than opportunistic, and the sections show a body that starts near surface and continues. The balance sheet is unlevered and the register is tight. The board has done this before, together, at Genesis, and the Managing Director is drilling a project he bought once already and never thought was finished. What argues against it. There is no resource, so there is nothing to value and nothing we would value if there were. The company owns 80.3 per cent of the asset, not all of it. The high-grade intercepts are nuggety and short, and the resource will be top-cut. The intervals are downhole lengths and true widths are unknown. A capital raising during 2027 should be assumed. The open pit is a stated potential, not a study. And the only processing solution that makes near-term development plausible depends on an arrangement with a third party that nobody has announced. We searched Ramelius' ASX announcements, its Mt Magnet Hub and Eridanus project pages and Ordell's own announcements on 2 September 2026, and found no reference to ore purchase or toll treatment between the two companies. Absence of a public reference is not proof that no conversation exists, and it is all we can report. On balance the Samso Call is TIME TO LOOK AT TAKING A POSITION. That call describes attention, not action. It says the sequence of events between now and the December quarter is worth watching closely and that the work required to form a view is worth doing now rather than after the announcement. It is not a recommendation to buy, it does not carry a price, and it does not carry a valuation. Samso publishes neither. The thing we would say to anyone reading this before the maiden resource lands is the thing this piece has tried to say throughout. The evidence Ordell has published is good enough to justify a resource estimate. It is not yet good enough to tell you how big that resource will be. Those are different sentences, and a great deal of money in the junior gold sector is lost by people who read the first one and hear the second. 17 Disclaimer The information contained on this website is the writer's personal opinion and is provided to you for information only and is not intended to or nor will it create/induce the creation of any binding legal relations. There may be a conflict of interest present with commercial arrangements with companies and/or stock held. Samso or an associate may receive a commission for funds raised. The information or opinions provided herein do not constitute investment advice, an offer or solicitation to subscribe for, purchase or sell the investment product(s) mentioned herein. It does not take into consideration, nor have any regard to your specific investment objectives, financial situation, risk profile, tax position and particular, or unique needs and constraints. Accordingly, no warranty whatsoever is given, and no liability whatsoever is accepted for any loss arising, whether directly or indirectly as a result of this information. Investments are subject to investment risks, including possible loss of the principal amount invested. The value of the product and the income from them may fall as well as rise. You may wish to seek advice from an independent financial adviser before committing to purchase or invest in the investment product(s) mentioned herein. The Samso Way – Seek the Research Here at Samso, we pride ourselves on delivering content for investors that is independent and informed by over three decades of experience in the industry. Our content is well-researched and is only created if I see merit in discussing the company's story. Our mission is simple. Cut through the noise and spotlight what matters. Genuine stories, grounded insights, and real opportunity. Investors can explore our four core platforms: Coffee with Samso Samso Insights Samso News Samso Research There may be numerous paths to success in investing, but the common thread among successful individuals is that they remain committed to making informed decisions. Equip yourself with the right knowledge and tools, and you will be well on your way to achieving your financial goals. The Samso Philosophy: Stay curious. Stay sharp. And remember, digging deeper always uncovers the real value. Happy Investing, and the only four-letter word you need to know is DYOR. Disclaimer Disclosure of interest. Samso Research Media House holds a small shareholding in Ordell Minerals Limited (ASX: ORD). This workup was not commissioned by Ordell Minerals and Samso received no payment from the company for producing it. This Samso Research Workup is independent research commentary for general information only. Nothing in this document is financial product advice, and it does not take into account any reader's objectives, financial situation or needs. Figures are drawn from public sources believed reliable at the stated dates but are not guaranteed; market-sensitive numbers change daily. Seek professional advice before making investment decisions. Read full Disclaimer. About Samso Samso is a trusted platform that equips dedicated investors with up-to-date industry knowledge and insights from top CEOs and thought leaders. Samso Research | www.samso.com.au | An Investor Lens on ASX-Listed Companies

Abstract Kaoko Metals (ASX: KAO) listed on 7 May 2026 at 20 cents. On 2 September 2026 it closed at A$1.85, up 151.7% in one session, after reporting visual copper mineralisation in the first two diamond drill holes ever put into its Chalkos Copper-Silver Project in north west Namibia. DDOT002 logged 60.25 metres of visible copper from 36.65 metres down hole. DDOT001 logged 51.83 metres from 39.27 metres. Neither number is a grade. Both are estimates made by eye, and the company says so on almost every page of its own announcement. The shares went on to close at A$2.50 on 3 September and were then halted. On 7 September the company announced a A$20 million placement at A$2.20. What has been established so far is narrow. Copper minerals are present, in core, at shallow depth, across several rock types, on ground that had never been drilled. Chalcocite, malachite, cuprite and native copper were all logged. Zones of mylonite sit alongside the strongest intervals, which supports the company's reading that structure as well as stratigraphy has focused the copper. That is a coherent geological story on a licence package of 39,033 hectares, one half of which has no recorded exploration of any kind. The belt carries a testable metal ratio. The Kaoko Belt is a copper-silver belt rather than the copper-cobalt belt of Zambia and the Congo, and published work in the same district reports 0.4% to 5.2% copper with 23 to 312g/t silver. Okohongo, elsewhere in the same belt, carries an Inferred Resource of 7.7Mt at 1.55% copper and 26.77g/t silver. Where a copper grade and a silver grade in this belt have been reported for the same material, the ratio runs between roughly 14 and 23 grams of silver for every 1% copper. The Omatapati figures above are two separate ranges rather than paired assays, so no ratio can be taken from them. The window gives a reader a benchmark to check the pending assays against, and it is set out in Section 04. Almost everything else remains unestablished. There is no Mineral Resource on either project, of any category. There is no economic study. True widths are unknown and both completed holes were collared from the same pad, so continuity is a hypothesis rather than an observation. The 69.6% copper headline grade comes from a hand-picked surface sample, one of just 35 ever taken, all of them on one of the two licences. Both Chalkos licences carried an expiry date of 30 June 2026 and renewal applications were still described as in process in September. The market has repriced the company by roughly twelve times since listing, on announcements containing no laboratory data. The company's own September presentation puts its market capitalisation at listing at A$12.1 million. On all shares on issue after the placement, not just the shares that are quoted, it is capitalised at approximately A$174 million at the 3 September close, against more than A$24 million of cash. Samso puts no price on this company and could not defend one if we did. There is no dataset to build it from. The placement is the second piece of evidence to arrive, and it is not a drilling result. It was priced at a 12.0% discount to the last close and a 20.2% premium to the fifteen-day average of A$1.83. The company reports new and existing long-only institutions and sophisticated investors in the book, without breaking it down. Cash goes from A$5.2 million to over A$24 million. A second diamond rig is named in the use of funds. Existing holders are diluted 13.0%. The placement shows how professional buyers priced the same incomplete information a reader is weighing. It says nothing about the grade the laboratory will report. The argument is settled by one event. Assays from DDOT001 and DDOT002 are expected within four to six weeks of 2 September 2026. Visual logging often fails to survive the laboratory, and where that happens the share price correction is usually quick. Everything in this workup is provisional until those numbers land. The call is WATCH & DYOR. The ground is scarce and the result is interesting. The price already assumes an answer to a question that has not been asked yet, so the four weeks before the assays arrive are the time to do the research. Contents 1 Why We Are Doing a Full Workup on This One 7 Valuation, Our Own Working 2 The Company and the Story 8 What Would Change Our Mind 3 The Assets 9 Management and DNA 4 The Kaoko Belt as a Copper-Silver System 10 Samso Concluding Comments 5 The Macro and the Commodity Backdrop 11 Disclaimer 6 How Kaoko Compares With Its Peers Snapshot Prices below are as at the date in the meta line above, the point this workup was compiled from. This is a living document. The share price and market cap will have moved, up or down, in the time since (check a live quote before acting on anything here, and see the note on data currency in the disclaimer). Metric Value Share price A$2.50 (last ASX close, 3 September 2026, then halted) Placement A$20.0m at A$2.20, announced 7 September 2026 Placement pricing 12.0% discount to the last close, 20.2% premium to the 15-day VWAP of A$1.83 Range since listing A$0.330 to A$2.62 (listed 7 May 2026 at A$0.20) Shares on issue 60,615,000, rising to 69,705,910 on allotment Market cap, all shares ~A$174.3m post-placement, at A$2.50 Dilution 13.0% of the enlarged capital Shares in ASX escrow 24,822,505, or 35.6% of the enlarged capital Cash Over A$24m after the placement, from A$5.228m at 30 June 2026 Mineral Resource None, on either project Assays DDOT001 and DDOT002 pending, four to six weeks from 2 September 2026 Broker coverage None Projects Chalkos (100%), Kunene Region. Karibib (earning up to 85%), Erongo Region Top 20 shareholders hold 42.97% of shares on issue (Market Index, current). The largest holders are Gerard O'Donovan 3,500,000 (5.77%), Russell Brooks Limited 2,510,055 (4.14%), Jurie Hendrik Wessels 1,885,436 (3.11%), Philip Le Roux 1,885,435 (3.11%), Megan Elizabeth Hamel 1,520,000 (2.51%). No substantial holder notices are recorded in the Market Index database. Percentages are quoted against all shares on issue, not against quoted capital, and they are struck before the placement. Every holding above is diluted by 13.0% once the new shares are allotted, unless the holder took part in the placement. Share price context. Kaoko traded between roughly 70 and 80 cents through late August, on volumes averaging under 330,000 shares a day. On 2 September it opened at A$1.30, traded as high as A$2.28, closed at A$1.85 and turned over 5.06 million shares worth A$9.06 million. That is more than fifteen times its four-week average volume in a single session. The following day it closed at A$2.50, and the shares were then halted pending the placement announced on 7 September. The stock is up more than twelve times its 20 cent issue price in four months. Every one of those moves rests on exploration news, none of it yet supported by a laboratory assay. 1 WHY WE ARE DOING A FULL WORKUP ON THIS ONE Kaoko Metals listed on 7 May 2026 at 20 cents. On 2 September 2026 it closed at A$1.85, up 151.7% in a single session, on an announcement that contained no assay results. Those two facts together are why we have written this workup. What the company announced was visual copper mineralisation. Its geologists looked at the first two holes ever drilled at the Chalkos Copper-Silver Project and logged what they could see in the core. In DDOT002 they logged copper minerals over 60.25 metres. In DDOT001 they logged them over 51.83 metres. Neither number is a grade. Both are estimates made by eye, and the company said so, repeatedly and in bold, in the same document. Samso does not usually write at this length about a company with no Mineral Resource, no economic study and no assays. We are doing it here for three reasons. The first is that a first-ever drill programme intersecting broad, shallow copper mineralisation across a licence package that had never been drilled is a real geological event, whatever the grade turns out to be. The second is that the market has already repriced the company by more than 150% on information that is explicitly incomplete. When a share price moves that far ahead of the evidence, the useful work is not to argue with the move. It is to set out precisely what is known, what is not, and what has to arrive for the price to be justified or unwound. The third is that the underlying land position is genuinely underexplored in a way that is now rare. Before Kaoko, the entire recorded exploration history of the Chalkos Project amounted to 35 grab samples on one of its two licences. The other licence has no recorded exploration at all. That is unusual, and it cuts both ways. Samso has not spoken with Kaoko Metals management in preparing this piece, and Kaoko Metals has not commissioned it, paid for it or seen it before publication. Everything here is drawn from the company's own public filings and from market data, both cited. 2 KAOKO METALS LIMITED - THE COMPANY AND THE STORY Kaoko Metals Limited is a Perth-based exploration company with two projects in Namibia. It was incorporated on 13 June 2025, converted to a public company on 6 February 2026, and listed on the ASX on 7 May 2026 after raising A$6.5 million at 20 cents a share. The IPO closed oversubscribed at maximum subscription. Figure 1. Kaoko Metals project locations. Chalkos is in the Kunene Region in Namibia's north west, roughly 400km west of Tsumeb and at about the same latitude. Karibib sits in the Erongo Region, roughly 130km west north west of Windhoek. The pink bands are the copper belts. Note the Midas Minerals (ASX: MM1) reference on the map, which is Kaoko's nearest listed analogue. Source: Kaoko Metals, ASX announcement 2 September 2026, Figure 7. The two assets Chalkos Copper-Silver Project sits in the Kaoko Belt in north west Namibia, held through two Exclusive Prospecting Licences, EPL 7608 and EPL 7943. An Exclusive Prospecting Licence, or EPL, is Namibia's exploration tenure. It gives the holder the sole right to explore a defined area for a fixed term. Kaoko owns 100% of both. Combined area is 39,033 hectares, which is 390 square kilometres. The company describes Chalkos as an 800 square kilometre project, and that figure does not reconcile with its own tenement schedule. See the discrepancies noted at the end of this section. Karibib Gold-Copper-Tungsten Project sits in the Damara Belt in central Namibia, held through EPL 4663, covering 24,960 hectares. Kaoko does not own it. It holds a staged right to earn in, and on admission to the ASX it held no ownership interest at all. The structure is layered. Kaoko can earn up to 100% of Karibib Pegmatite Exploration, which holds 85% of Goas Pegmatite Exploration, which holds the licence. The effective ceiling is an 85% interest in the project. Where the money is Chalkos is the priority and the budget says so. The prospectus allocated A$1.5 million to A$1.87 million to Chalkos exploration in year one and A$1.48 million to A$1.78 million in year two. Karibib gets A$200,000 in year one and A$300,000 in year two, against A$400,000 of earn-in fees across the same period. On the company's own numbers, Karibib costs almost as much to hold as it does to explore. After the June 2026 site visit the company reordered its own plan. Karibib was to be drilled first. Chalkos was moved ahead of it. The first two holes have since returned visible copper, though the decision itself was made on the strength of surface mapping rather than drilling, and the assays that would test it are still outstanding. Market capitalisation, quoted shares against all shares Most price screens will tell you Kaoko is a much smaller company than it is. That figure is calculated on quoted shares only. It excludes the 24.8 million shares held in ASX escrow, which are real shares with real economic ownership. They are simply not tradeable yet. On all shares on issue the market capitalisation at the 3 September close of A$2.50 is approximately A$151.5 million, and the company uses the same figure in its September investor presentation. After the placement shares are allotted the same price gives approximately A$174.3 million. When you are comparing Kaoko to anything else, that is the number to use. Escrow falls from roughly 41% of the equity to 35.6% on the enlarged capital, and the first tranche, 11.57 million shares, comes free 12 months from issue. Anyone modelling the register needs that date in the calendar. The overhang Beyond the escrow there are 4.5 million options struck between 30 cents and 50 cents, all now deeply in the money at A$2.50, and 5.55 million performance rights issued to the vendors of Chalkos. The prospectus text we have been able to read does not set out the vesting conditions on those rights, which sit in a section of the document that was not available to us. That is a genuine gap and we say so rather than guessing. It is a reasonable question to put to the company. On a fully diluted basis the securities on issue at listing totalled 70,665,000. Adding the placement shares and the 2.25 million shares issuable under the Karibib earn-in takes the fully diluted count to 82,005,910. The placement, and what it changes On 7 September 2026 the company announced firm commitments for a A$20.0 million placement of 9,090,910 shares at A$2.20. Euroz Hartleys and Cumulus Wealth were joint lead managers and bookrunners. Settlement is expected on 14 September and the new shares are expected to begin trading on 15 September. No shareholder approval is needed. The issue uses 6,061,500 shares of placement capacity under ASX Listing Rule 7.1A and 3,029,410 under Listing Rule 7.1. The issue price is a 12.0% discount to the last close of A$2.50 on 3 September. It is also a 20.2% premium to the volume weighted average price of A$1.83 over the fifteen trading days to that date. The issue price is above the fifteen-day average and below the last close. Both comparisons are set out here, and a reader can weigh which is the fairer reference. The company says the placement was supported by new and existing long-only global institutions and by sophisticated investors. Samso's view is that the placement is the most useful piece of external evidence to arrive since listing. It is not a drilling result. Institutional, professional and sophisticated investors have put A$20 million behind core photographs and a belt argument, at a price a fifth above where the stock had been averaging, before a single grade has been reported. The company does not break the book down, so how much of it is institutional is not on the record. That does not show the copper is economic. It shows how a set of professional buyers priced the same incomplete information a reader of this workup is weighing. The placement changes the balance sheet more than it changes the share price. Cash goes from A$5.228 million at 30 June to more than A$24 million before offer costs. On the June quarter's outgoings of A$775,000 that is years of funding rather than quarters, and it pushes the financing question well beyond the current programme. The company's stated use of funds is a second diamond rig at Chalkos, an expanded programme at Otniel and Donkey Hill, regional mapping, sampling and geophysics across the wider licence area, and geophysics at Karibib ahead of a maiden drill programme. The announcement calls that allocation a statement of current intentions that may change, and the presentation describes the second rig as still under investigation. The cost is dilution of 13.0%. A holder who did not take part now owns 87% of the stake they held before, in a company with more than four times the cash and a larger drilling programme. Whether that trade was worth taking is a judgement, and it depends entirely on the assays. Sources: Kaoko Metals, ASX announcement 7 September 2026. Kaoko Metals investor presentation, September 2026. 3 THE ASSETS Chalkos, and what the first two holes showed On 2 September 2026 Kaoko reported the results of the first two diamond drill holes ever completed at Chalkos, both at the Otniel prospect. Diamond drilling recovers a solid cylinder of rock, called core, which can be logged, photographed and split for assay. It is the highest-fidelity way to sample a rock at depth, and it is more expensive and slower than the alternatives. Figure 2. Drill hole location plan at Otniel. DDOT001 and DDOT002 were collared from the same pad, marked by the single orange dot. DDOT003 tests along strike, DDOT004 to the west and DDOT005 to the south. Note the mapped anticlines and synclines in blue, which is the fold architecture the company believes is helping to localise the copper. Source: Kaoko Metals, ASX announcement 2 September 2026, Figure 3. Both holes were collared from the same pad. The company says this reflected site topography, water access and the state of the earthworks at the time, which had not yet caught up with the drill. That is a practical constraint rather than a design choice, and it limits how much the two holes can say about lateral continuity. They are, in effect, two angles from one point. Figure 3. Cross section through DDOT001 and DDOT002 showing the logged intervals of visible copper mineralisation against schematic geology. The orange bars are visible mineralisation and the red bars are what the company calls strong visible mineralisation, meaning intervals where the estimated copper minerals are typically 5% or more by volume. The geology behind the holes is interpretation, not measurement. Source: Kaoko Metals, ASX announcement 2 September 2026, Figure 4. Figure 4. Core from DDOT002 at 68.13m to 68.34m on the left, and DDOT001 at 64.03m to 64.17m on the right. In the left panel the company logs malachite, chrysocolla, cuprite and chalcocite. In the right panel it logs chalcocite, cuprite and dioptase, so the green there is dioptase rather than malachite. This is what the market repriced on, and it is a photograph, not an assay. Source: Kaoko Metals, ASX announcement 2 September 2026, Figures 1 and 2. The Managing Director put the result this way on the day of the announcement. "While it is still early in the program and we are waiting for results of the initial assays, we are pleased to have observed broad zones of copper mineralisation in the first two holes at Otniel. DDOT002 intersected over 60m of predominantly malachite-chalcocite mineralisation, including a 32m zone of strong visible mineralisation, while DDOT001 returned over 51m of visible mineralisation including 17m of strong visible mineralisation." "This is a highly encouraging start to the first ever drilling of the 800km2 Chalkos Project, and we are excited to continue drilling at both the Otniel and Donkey Hill prospects." Gerard O'Donovan, Managing Director, Kaoko Metals, ASX announcement, 2 September 2026 That statement is carefully built. It leads with the caveat, it uses the word observed rather than intersected for the mineralisation itself, and it does not put a grade on anything. Read against the announcement it accompanies, it claims no more than the logging supports. Two things in it are worth carrying forward. The 800 square kilometre figure does not match the company's own tenement schedule, which is discussed under discrepancies below. And the word encouraging is doing the work that the assays have not yet done. How to read the result There are four things in this announcement that a careful reader should weight, and two of them are genuinely encouraging. The first encouraging point is depth. Continuous visible mineralisation starts at 36.65 metres in DDOT002 and 39.27 metres in DDOT001. That is shallow, and depth matters to the economics. If this system ever becomes a mine, mineralisation starting at 40 metres implies an open pit and mineralisation starting at 400 metres implies an underground operation. The second is that the mineralisation is not confined to one rock type. It is logged across quartzite, conglomerate, siltstone and dolostone. Both holes also record zones of mylonite, which is rock that has been ground and stretched by deformation, sitting coincident with or immediately next to the strongest mineralisation. The company reads this as evidence that structure, meaning faults and shearing, helped focus the copper-bearing fluids alongside the favourable rock units. If that reading holds, the system is not restricted to a single bed and there is more room to hunt. The third point is the one the company repeats in bold through its announcement. These are visual estimates. A geologist looking at a core and calling 6% malachite is estimating volume by eye. Malachite is about 57.5% copper by weight, chalcocite about 79.8%, cuprite about 88.8%. It is arithmetically tempting to turn a volume percentage into a grade. The company explicitly forbids it, and it is right to. Recovery, density, the proportion of the rock actually sampled and the accuracy of the eye all sit between the observation and the number. Assays are expected within four to six weeks of 2 September 2026. The fourth is true width. Both holes report down-hole lengths only. The relationship between the drill orientation and the geometry of the mineralised zone is not yet established, and the company says so. A 60 metre down-hole interval could be a 60 metre true thickness or a much narrower zone cut at an unfavourable angle. Two holes from one collar cannot resolve that. The distance from here to a Mineral Resource is large. The JORC Code prescribes no drill spacing for any confidence category. Classification is the judgement of the Competent Person, based on whether the data establish geological and grade continuity. Two completed holes from a single collar, with assays outstanding, cannot support a resource of any category, and the company has not suggested otherwise. The three holes that follow, from three separate collars, are the first that could begin to establish continuity. "Visual estimates of mineral abundance or type should never be considered a proxy or substitute for laboratory analyses. Laboratory chemical assays are required to determine definitive mineral types, concentrations, or grades, which are the factors of principal economic interest." Kaoko Metals, ASX announcement, 2 September 2026 What was known at Chalkos before the drill arrived Very little had been done, and that matters to the argument in both directions. The prospectus states that the only recorded exploration on the Chalkos Project was mapping and soil sampling by LexRox Exploration on EPL 7943. That work produced 35 grab samples in total across two phases. A grab sample is a hand-picked piece of rock. It is chosen because it looks interesting, which makes it useful for confirming that something is present and close to useless as a measure of average grade. Those grab samples returned up to 52.7% copper at Otniel, up to 69.6% copper at Donkey Hill, and up to 31.97% copper in a later phase. The 69.6% figure appears on the front of the company's presentations. It is a correctly reported number, and it is a peak value from a hand-picked sample rather than an average. On EPL 7608, which is 19,951 hectares of the 39,033 hectare project, the prospectus states that no exploration has been recorded at all. Metallurgy has been tested at a preliminary level. XRT ore sorting, which separates rock by X-ray transmission, upgraded a 308kg Donkey Hill bulk sample from a 9.9% copper head grade to 13% copper, at a yield of 64% and a copper recovery of 82%. A 208kg Otniel sample went from a head grade the prospectus gives as either 3.0% or 3.5% copper to 8% copper, at a yield of 35% and a copper recovery of 61%. The recovery figures carry as much weight as the upgrade. Any sample can be upgraded by rejecting enough of it, so the copper lost to the reject stream is the number that tells a reader what the method costs. The Otniel figures do not reconcile with each other. At the stated yield and recovery the product would grade about 6% copper rather than 8%. The Donkey Hill set does reconcile. Both are quoted as published and the reconciliation is a question for the company. Acid leach testwork on samples grading 7.91% and 10.06% copper recovered 89% from one and 71.7% from the other. The company noted that the second sample showed high acid consumption, which it described as potentially prohibitive for commercial processing without treatment of the gangue, meaning the non-copper-bearing rock. That caveat sits in the prospectus and does not appear in the presentation. Two limits on that testwork are worth stating, because neither is obvious from the headline recoveries. The mineralogy reported alongside it, in the company's investor presentation of 7 May 2026, is oxide, with brochantite, malachite and spangolite identified by X-ray diffraction, so what has been tested is weathered material. We can find no published results for primary sulphide in the prospectus or in any announcement to 2 September 2026, and sulphide is what a deposit of any depth would eventually have to treat. The second limit is grade. The head grades tested run from about 3% to 10% copper. Okohongo's Inferred Resource in the same belt grades 1.55% copper. Recoveries measured on very high grade material do not tell a reader how ordinary resource-grade rock will behave, and the leach times behind these figures are not stated in the pages available to us. The size fractions are stated. The prospectus records the material crushed and screened to a p80 of minus 25mm, and the column work run on minus 2mm Otniel material. What the September presentation adds at surface The investor presentation released with the placement carries surface work that the 2 September drilling announcement did not. The company reports a new prospect at Chalkos called Bootless, where outcropping copper mineralisation has been mapped. It reports the mapped footprint at Otniel extended by up to 200 metres, taking it beyond 600 metres across multiple trends. At Donkey Hill it reports an extension of about 60 metres at Donkey Hill East, taking that footprint beyond 800 metres across three trends. It also describes 20 kilometres of confirmed strike on the belt with a further 20 kilometres of untested prospective ground along the same trend. Every one of those figures is a mapped surface footprint or a company estimate, not drilling and not an assay. The same cautionary note the company applies to its core applies here. They do change the size of the target list. Donkey Hill carries the 69.6% surface grab sample and has not been drilled at all, and the company has now funded the rig time to test it. The presentation gives the maiden programme as roughly 3,000 metres, already expanded at Otniel, and says the company is investigating the mobilisation of another diamond rig. The area figures have moved, and the disclosure is now better than it was. The placement announcement describes the Chalkos licence area as approximately 400 square kilometres. That does reconcile with the 39,033 hectares in the tenement schedule, which is 390 square kilometres. The 800 square kilometre figure set out at the end of Section 02 has not been repeated. The presentation still carries 80,000 hectares in one place, which is the old 800 square kilometres in a different unit. So the company has corrected the figure in the announcement and not in the presentation. Both are reported here as published. One statement in the presentation needs reading against Section 02. Chalkos is described as fully permitted for immediate drilling. Both Chalkos licences carried an expiry date of 30 June 2026 and the renewal applications were still described as in process in September. Both can be true at once, and the two belong side by side. Sources: Kaoko Metals investor presentation, September 2026. Kaoko Metals, ASX announcement 7 September 2026, for the approximately 400 square kilometre licence area. Kaoko Metals Prospectus dated 23 February 2026 for the tenement schedule and the Donkey Hill grab sample. Karibib Figure 5. The Karibib Project, EPL 4663, in the Erongo Region. The prospects sit along a north east to south west structural corridor described by the company as 20km long and 2km wide. Source: Kaoko Metals, ASX announcement 16 July 2026, Figure 1. Karibib is a different kind of asset and a different kind of risk. It is a skarn and replacement system, meaning mineralisation formed where hot fluids from an intrusion reacted with carbonate rocks. It carries copper, gold, silver and tungsten. It sits in the Damara Belt, roughly 32km from the Navachab gold mine and about 40km from the Twin Hills deposit. In May 2026 Kaoko took 54 rock chip samples across three prospects. Ten came back above 1% copper, three above 2% copper, and eleven above 0.1% tungsten trioxide. The strongest single result was sample Z2809 at 3.19% copper, 0.85g/t gold, 46g/t silver and 0.48% tungsten trioxide. Historical work at Karibib is more extensive than at Chalkos, and the results were less encouraging. Gold Fields held ground there in 1983 and 1984 and let the licence lapse, having concluded the results did not warrant follow-up. Anglo American held it in 1988 and 1989, identified three areas with average gold values between 2 and 7 parts per billion, never followed them up, and allowed its licence to expire in 1990. The prospectus states plainly that the coordinates of both companies' sampling are unknown. More recently, Arcadia Minerals drilled 10 reverse circulation holes for 551 metres in 2022. Reverse circulation drilling returns rock chips rather than solid core. It is faster and cheaper than diamond drilling and less precise. The two best intercepts were 4 metres at 1.35% copper and 0.68g/t gold from 24 metres, and 4 metres at 1.98% copper, 0.92g/t gold and 0.72% tungsten from 9 metres, reported by Arcadia as tungsten metal rather than as the tungsten trioxide used in Table 4. These are the numbers that appear in Kaoko's marketing. They are correctly reported, and both intercepts are only 4 metres wide. The 2022 drilling was not done by Kaoko and no drilling has been done at Karibib since. The company has instead started a systematic regional geochemical programme, sampling soil and calcrete across the licence, which is the method Osino Resources used to find Twin Hills. That is a sensible sequencing decision. It is also two to three years of work before a drill target list of any confidence exists. Tenure Both Chalkos licences carried an expiry date of 30 June 2026. The company has submitted renewal applications for a further two-year term and told the market in September 2026 that it expects them to be processed in line with standard regulatory timelines. As at the date of this workup we have found no announcement confirming the renewals as granted, having read every Kaoko release listed on Market Index from admission on 6 May 2026 to 2 September 2026. Both Chalkos licences depend on that renewal. On the public record the position is pending, and resolving it sits with the company and the Namibian Ministry of Mines and Energy. The prospectus also disclosed, in its risk section, that certain licence conditions attaching to the EPLs comprising the projects have not been strictly complied with, and that under the Namibian Minerals Act the Minister may cancel a licence where the holder fails to comply. The company said it was not aware of any such notice and considered the likelihood of forfeiture low. The underlying detail sits in the Title Report annexed to the prospectus. At Karibib there is a separate access problem. The land use agreement for Farm Gamikaub terminated on 9 May 2021. Access is currently granted on an ad hoc basis by the landowner. The prospectus itself says this increases uncertainty and may disrupt systematic exploration unless the agreement is renewed or replaced. There is also a live policy question. Namibia has a draft Minerals Bill contemplating mandatory local content and minimum Namibian equity participation. Earlier drafts referenced a 5% minimum. The prospectus notes that public discussion has included reference to potential requirements of up to 51% local ownership, and that if such requirements are imposed the company may be required to divest or dilute its interest in the projects. That is the company's own disclosure, not our speculation. Discrepancies in the record Two figures in the company's own documents do not reconcile, and Samso's practice is to state what is published, name the discrepancy and move on. Resolving it belongs to the company. The prospectus refers in one place to a "large 80,000Ha land package". The tenement schedule in the same document lists 19,951 hectares plus 19,082 hectares for the two Chalkos licences, and 24,960 hectares for Karibib, totalling 63,993 hectares across all three. The same gap appears in the announcements in different units. Kaoko describes Chalkos as an 800 square kilometre project. The two Chalkos licences total 39,033 hectares, which is 390 square kilometres. Both the 80,000 hectare and the 800 square kilometre figures are close to double the areas listed in the tenement schedule. Separately, the same Otniel bulk metallurgical sample is given a head grade of 3.0% copper in one paragraph of the prospectus and 3.5% copper two paragraphs later. 4 THE KAOKO BELT AS A COPPER-SILVER SYSTEM Kaoko's independent geologist notes in the prospectus that the Kaoko Belt shares similarities with the Central African Copperbelt but is not an exact analogue. One of the differences he lists is the metal association. The Zambian and Congolese deposits are dominantly copper-cobalt. The Kaoko Belt is dominantly copper-silver. He presents that as a caveat. It also tells us what kind of deposit Chalkos is being tested for, and it gives us a metal ratio to check the pending assays against. Figure 6. The Kaoko Belt in its regional setting. Chalkos sits in this belt, in the Kunene Region, roughly 43km from Sesfontein township. The belt is cut by named regional mylonite zones, the Purros (PMZ) and Three Palm (TPMZ) Cambrian shear zones, and by the Sesfontein Thrust (ST). That structural architecture is the context for the mylonite Kaoko logged in DDOT001 and DDOT002. The inset at top right places the Kaoko Belt alongside the Ghanzi-Chobe, Katanga and Zambezi belts, which is the Central African Copperbelt comparison the story rests on. Reproduced from Nghoongoloka et al. (2024), Resource Geology 74(1) e12341, Figure 1, modified after Hoffmann and Prave (1996), Goscombe et al. (2003), Cailteux et al. (2005), Gray et al. (2006), Miller (2008) and Webb (2013). What a sediment-hosted copper deposit is Chalkos is being explored as a sediment-hosted stratiform copper system, usually shortened to SSC. These are not deposits formed by a volcano or an intrusion pushing metal up from below. They form inside sedimentary basins, over long periods, and the mechanism is chemical rather than igneous. The standard account is set out by Hitzman and co-authors in the Economic Geology centenary volume. Oxidised, salty groundwater moves through red sandstones and siltstones. Those red beds are red because the iron in them is oxidised, and oxidised brine passing through them strips out copper and carries it in solution. The brine keeps moving until it meets rock that is chemically reducing, typically a dark, organic-rich or pyritic shale or siltstone. At that boundary the chemistry flips, the copper drops out of solution, and it precipitates as sulphide. Geologists call that boundary a redox front. Squire and Keays put the significance of these systems plainly. Sedimentary rock-hosted stratiform copper deposits are the world's second largest source of copper and the largest source of cobalt, and roughly 73% of that copper sits in just two basins, the Katangan Basin of the Central African Copperbelt and the Permian Basin that hosts Europe's Kupferschiefer. Their description of the process is the same one. Oxidised brines flush through the basin, and where the liberated copper reaches carbon-rich mudstone, the metals are deposited. This matters for Kaoko because the company's own model at Chalkos is exactly that. In the JORC table appended to its 2 September 2026 announcement, the company describes copper concentrated in a dark grey to grey green phyllitic siltstone and shale, interbedded with dolomite of the Lower Omao Formation, sitting between the red beds of the Nosib Group below and the Omao carbonates above. It calls that unit the Chalkos Horizon and interprets it as the first and lowest redox boundary within the stratigraphic column. That is a standard description of a sediment-hosted copper target. The United States Geological Survey published a formal deposit model for this class in 2015, by Hayes and co-authors, and it puts numbers on the fluid. The ore-forming brines were warm, between roughly 75 and 220 degrees Celsius, oxidised, and very saline, typically carrying more than 15 per cent sodium chloride equivalent. They moved upward across the bedding toward the host rocks. The reductant that stopped them was solid organic matter, petroleum or sour gas, or carbonised plant fragments, depending on the deposit subtype. Figure 7. How a sediment-hosted stratiform copper system forms, drawn onto the stratigraphy Kaoko is targeting at Chalkos. The sequence is the same everywhere this deposit type occurs. An oxidised, saline brine strips copper from red beds, travels up-dip through permeable rock, and dumps its copper as sulphide where it meets a reducing horizon. Step three, the weathering overprint, happens hundreds of millions of years later and is discussed further down this section. Samso Research Media House original figure, built from the ore-system model of Hitzman et al. (2005) and Hayes et al. (2015), with stratigraphic names as used by Kaoko Metals. It is a schematic and is not drawn to scale. Nothing about the geometry at Chalkos has been established by drilling. Figure 8. The same architecture, published, from a prospect in the same belt. Basement at the base, then the Nosib Group, then the Beesvlakte and Devede Formations, with the stratiform copper marked in red at the top of the sequence. Compare the stacking order with Figure 7 above and with the Chalkos Horizon description in Kaoko's prospectus. The core photograph at right is massive chalcocite with malachite from 37 metres down hole, which is the same mineral pair and roughly the same depth Kaoko logged at Otniel. Reproduced from Nghoongoloka et al. (2024), Resource Geology 74(1) e12341, Figure 2b, modified after Bell (2012) and Webb (2013). The metal that comes with the copper Sediment-hosted copper systems are not all the same and the difference shows up in what rides along with the copper. The Central African Copperbelt carries cobalt, and Cailteux and co-authors document that copper-cobalt character across the Congolese and Zambian deposits. The Kupferschiefer carries silver, and it carries a great deal of it. The USGS deposit model sets out the pattern behind that. Hayes and co-authors record that 25 per cent of these deposits contain silver, grading between 2 and 200 grams per tonne, and 14 per cent contain cobalt, grading between 0.05 and 0.5 per cent in primary ore. They also record that silver is nearly totally confined to the zones bearing chalcocite and bornite, while cobalt extends into the pyritic zones. Deposits that carry by-product silver do not carry by-product cobalt, and the reverse holds as well, with the Kupferschiefer the only stated exception. The Kaoko Belt being copper-silver rather than copper-cobalt therefore says something about which part of the mineral zonation a deposit occupies. Within this deposit class the two metal associations are close to mutually exclusive. Figure 9. The sulphide zonation in a sediment-hosted copper system, and where the by-product metals sit within it. The sequence runs pyrite, chalcopyrite, bornite, chalcocite outward from the redox front. Silver rides with the copper-rich end of that ladder. Cobalt rides with the iron-rich end. Kaoko logged abundant chalcocite and only trace chalcopyrite in DDOT001 and DDOT002, which places what it has drilled so far at the silver-bearing end. Samso Research Media House original figure. Zonation sequence and the silver and cobalt distributions after Hayes et al. (2015), USGS Scientific Investigations Report 2010-5070-M. The Kupferschiefer shows how large the silver side of such a system can be. The Polish Geological Institute reports Poland's anticipated economic copper resources in the Fore-Sudetic Monocline and North Sudetic Trough as 1,976 million tonnes of ore containing 36 million tonnes of copper and 107,000 tonnes of silver. That averages roughly 1.8% copper and roughly 54 grams per tonne silver. The same source records that since 2011 KGHM Polska Miedź has been the largest producer of silver worldwide. The world's largest silver producer is therefore a copper miner, and the rock it mines is a sediment-hosted copper-silver shale. Silver in these systems can be a co-product carrying a material share of the revenue. Exactly how the silver is held inside those copper minerals is still an active research question. Kelepile and co-authors examined that problem in the Kalahari Copper Belt, the other great Neoproterozoic copper-silver belt of southern Africa, in a 2020 paper on the partitioning and distribution of silver in sediment-hosted copper-silver deposits. We flag it as the reference work rather than summarising conclusions we have not read in full. For this workup the USGS statement is sufficient. The silver sits in the chalcocite and the bornite. Figure 10 is the published version of that zonation, showing the mineral zoning of the Spar Lake copper deposit in Montana as mapped by the USGS. The sulphide list runs down the left and the mineral zones run across the top. Figure 10. Mineral zones at Spar Lake. Stromeyerite, argentite and native silver appear in the red and orange columns, which are the chalcocite-chlorite and bornite-calcite zones. Cobaltite, cobalt-anomalous pyrite and nickel-rich pyrite appear in the blue and green columns, the galena-calcite and pyrite-calcite zones at the other end of the system. Silver sits at the copper-rich end and cobalt at the iron-rich end, within a single deposit. Stromeyerite is a copper-silver sulphide. Reproduced from Hayes, Cox, Piatak and Seal (2015), USGS Scientific Investigations Report 2010-5070-M, Figure 6, itself modified from Hayes and others (2012). The Kaoko Belt's own silver record Copper-silver mineralisation in the Kaoko Belt has been drilled, sampled and in one case estimated by other parties, on ground within a few hundred kilometres of Chalkos. That record is the closest evidence available for what Chalkos may contain. The most directly relevant published science is a 2024 paper by Nghoongoloka and co-authors in Resource Geology, on the genesis of copper and silver mineralisation at the Omatapati prospect in the Opuwo district of the Kunene region. That is the Kaoko Belt, and it is the same administrative district as Chalkos. The paper reports mineralisation hosted in Neoproterozoic dolomite and argillite, with ore grades ranging from 0.4 to 5.2 weight per cent copper and 23 to 312 grams per tonne silver. It describes three hypogene stages involving calcite-quartz-barite veins carrying chalcopyrite and bornite, and it concludes that Omatapati represents a sediment-hosted copper-silver deposit similar to those in the Central African Copperbelt. The counterweight belongs in the same paragraph. The authors report fluid inclusion temperatures of 145 to 300 degrees Celsius and interpret magmatic fluids mixed with meteoric water or seawater, and the mineralisation they describe sits in veins. That is a hotter and more hydrothermal picture than the low-temperature basinal brine model Kaoko applies at Chalkos. The paper supports the belt carrying copper-silver mineralisation of this family. It does not confirm that Chalkos formed by the same process. The stratigraphic position is similar at both prospects. The authors place the Omatapati mineralisation in the Devede Formation, sitting above the Beesvlakte Formation, which sits above the Nosib Group, which sits on basement. Kaoko puts its Chalkos Horizon in a comparable position, between the red beds of the Nosib Group below and carbonates above. The two descriptions were made independently. Further east, the Okohongo deposit has been drilled repeatedly. INV Metals reported hole INVR-037 returning 31 metres at 1.8% copper and 40.6g/t silver, and an earlier hole 6 returning 45 metres at 2.0% copper and 27.1g/t silver. In 2021 White Metal Resources filed a technical report with an Inferred Mineral Resource of 7,706,732 tonnes at 1.55% copper and 26.77 grams per tonne silver, at a 0.30% copper cut-off, containing 119,256 tonnes of copper and 6,634,133 ounces of silver. Midas Minerals reported its maiden Inferred Resource at Otavi as 10.5 million tonnes at 1.6% copper and 21 grams per tonne silver. One more piece of context before the tests. The USGS compiled grade distributions for 135 sediment-hosted copper deposits worldwide, and the silver one is reproduced below. It shows the proportion of deposits carrying more than a given silver grade. Figure 11. Silver grades in sediment-hosted copper deposits worldwide, from a population of 135. The curve starts at roughly 0.27. The curve starts at roughly 0.27, so only about a quarter of these deposits carry meaningful silver at all, which matches the 25 per cent the USGS deposit model states. And the marked line sits at 30 grams per tonne against a proportion of 0.10, so only one deposit in ten anywhere in the world exceeds 30 grams per tonne silver. Reproduced from Cox, Lindsey, Singer, Moring and Diggles (2003), Sediment-hosted copper deposits of the world, USGS Open-File Report 03-107, Figure 3. Against that distribution, Okohongo's Inferred Resource of 26.77 grams per tonne silver sits just under the line marking the global top tenth. Omatapati's reported range of 23 to 312 grams per tonne would put much of it beyond that line. On these published numbers the Kaoko Belt's silver tenor is high by world standards for this deposit type. Whether Chalkos does the same is not yet known. The 2 September 2026 announcement reports copper minerals only, with no silver in its logging table and no silver assay data, and the word silver appears in it solely in the project name and in boilerplate. So nothing Kaoko has published tells us whether its copper carries silver at Okohongo's tenor, at a fraction of it, or at all. We assume the core is being assayed for silver, because the project is named Copper-Silver and the belt carries it. That is an assumption rather than a fact. We also do not know what weight management places on silver, or whether it is treated as a target in its own right. What this gives a reader before the assays land Three things, and they are all testable within weeks. First, is there silver at all. If the DDOT001 and DDOT002 assays return copper with little or no silver, then whatever Chalkos is, it is behaving differently from the systems the belt has produced everywhere else it has been drilled. That would not make the copper worthless. It would materially weaken the belt-analogue argument that underpins the whole story, and it would remove a revenue stream the market is currently assuming. Second, the ratio. Where a copper grade and a silver grade in this belt have been reported for the same material, the ratio runs between roughly 14 and 23 grams of silver for every 1% copper. Okohongo's Inferred Resource is at about 17, and its two published intercepts are at about 23 and about 14. Omatapati is not in that count. Its copper and its silver are published as two separate ranges rather than as paired assays, so a ratio cannot be taken from them. Outside the belt, the Otavi resource is at about 13 and the Polish Kupferschiefer averages about 30. If Chalkos lands inside the belt window, the case that it is the same kind of system gets a great deal stronger, whatever the absolute grade. If it lands far outside it in either direction, that is a question worth asking the company. Third, whether the silver tracks the chalcocite. The USGS model says silver in these deposits is nearly totally confined to the chalcocite and bornite zones. Kaoko logged chalcocite at 5% to 8% through DDOT002's strongest interval, from 59.30 to 91.66 metres. That is a specific, checkable prediction. If the silver assays peak in those same metres, the mineralogy and the geochemistry are telling the same story and the deposit model is behaving. If the silver is somewhere else entirely, or absent where the chalcocite is thickest, something in the model does not fit and it is worth finding out what. A caution on the shallow copper The same literature carries a caution about core descriptions of this kind. Readers should understand that the results, and what they imply about the system, may still end up as a disappointment without being technically wrong. The Omatapati paper describes a supergene process producing a semi-massive chalcocite and covellite zone extending from surface to roughly 50 metres depth. Supergene means secondary, formed by weathering. Rainwater works down from surface, dissolves copper from the primary rock, carries it lower and redeposits it as a concentrated blanket. It creates a zone that is richer than the rock beneath it and shallower than the rock that made it. Kaoko logged malachite, azurite, chrysocolla, cuprite, native copper and chalcocite, from 36.65 metres to 96.90 metres down-hole in DDOT002. At dips of 55 and 65 degrees, that down-hole window is roughly 30 to 80 metres of true vertical depth, which is the depth range Omatapati's supergene blanket occupies. One point of care is needed here, because chalcocite by itself does not settle the question. Figure 10 shows chalcocite as the innermost primary zone of an unweathered deposit, and it is also the classic mineral of supergene enrichment. It occurs both ways. The other minerals logged alongside the chalcocite point toward weathering. Malachite, azurite and chrysocolla are copper carbonates and silicates formed at surface conditions, and cuprite and native copper belong to the same environment. That assemblage, at this depth, is the signature of a weathering profile rather than of the primary system. Figure 12. Supergene copper minerals under the microscope, from the Kaoko Belt. In panel (a) the basket weave textures are chalcocite, digenite, delafossite and hematite replacing chalcopyrite, which is the primary sulphide being broken down and reformed as secondary minerals. Panel (b) is semi-massive covellite. Panels (c) and (d) show stockwork chalcocite with malachite, goethite and hematite. Cct chalcocite, Dg digenite, Cv covellite, Ccp chalcopyrite, Mal malachite, Del delafossite, Hem hematite, Goe goethite, Brt barite, Qtz quartz, Wul wulfenite. Reproduced from Nghoongoloka et al. (2024), Resource Geology 74(1) e12341, Figure 10. The company itself points at this. It notes trace chalcopyrite within brecciated clasts in the shear fabric, and describes that as giving an indication of the primary mineralisation at depth. Chalcopyrite is the primary sulphide. Its scarcity in these holes, against abundant chalcocite and copper oxides, is consistent with the drill having tested the weathered upper part of a system rather than its core. Figure 12 shows that process under a microscope, from the same belt. In panel (a) the chalcopyrite is being replaced by chalcocite, digenite, delafossite and hematite. The copper in that rock was not deposited there by the ore-forming fluid. It was moved and reconcentrated upward by weathering. The Omatapati authors also report sulphur isotope values for the supergene chalcocite and covellite averaging plus 9.1 per mille, distinctly heavier than the primary sulphides in the same system, which is a second line of evidence that the shallow material has a different origin from the rock below it. None of this counts against the result. Supergene enrichment is real copper, it is shallow, and shallow material is generally cheaper to mine than deep material. Kaoko's own preliminary leach testwork was done on oxide material. It does limit what the first two holes can prove. A rich weathered blanket indicates that a copper system exists beneath it. It says much less about the size or grade of that system at 200 or 400 metres, which is the depth at which a deposit of any scale would have to sit. Structure and the belt's mylonite zones One last piece of belt geology. Both Kaoko holes record mylonite, meaning rock ground and stretched by ductile deformation, sitting coincident with or immediately next to the strongest visible mineralisation. The company reads that as structure helping to focus copper deposition. The regional literature makes that a reasonable reading rather than a hopeful one. Goscombe, Hand and Gray describe the Kaoko Belt as a classic transpressional orogen, meaning a belt built by simultaneous squeezing and sideways sliding, which is precisely the setting that generates pervasive ductile shear fabric. Figure 6 shows the scale of that deformation. The belt is divided into five tectonostratigraphic zones and the dividers are themselves named and mapped mylonite zones. Nghoongoloka and co-authors describe the Three Palm and Purros mylonite zones as Cambrian shear zones with an apparent age of 492 to 467 million years, alongside the Sesfontein Thrust. Chalkos sits within that architecture, so sheared rock in a Kaoko Belt drill hole is a common feature rather than a finding in itself. Reading it as a positive control on mineralisation remains a hypothesis that the assays will begin to test. Deformation can work in either direction. It can open pathways and concentrate metal, which is the reading the company offers. It can also smear, thin and dismember a stratiform layer that was originally continuous, which would help explain why a belt this prospective has produced so few deposits of size. Both readings are consistent with what has been logged so far, and only further drilling will separate them. References for this section Cailteux, J.L.H., Kampunzu, A.B., Lerouge, C., Kaputo, A.K. and Milesi, J.P., 2005. Genesis of sediment-hosted stratiform copper-cobalt deposits, central African Copperbelt. Journal of African Earth Sciences, 42, 134 to 158. DOI 10.1016/j.jafrearsci.2005.08.001. Goscombe, B., Hand, M. and Gray, D., 2003. Structure of the Kaoko Belt, Namibia, progressive evolution of a classic transpressional orogen. Journal of Structural Geology, 25, 1049 to 1081. DOI 10.1016/S0191-8141(02)00150-5. Hitzman, M., Kirkham, R., Broughton, D., Thorson, J. and Selley, D., 2005. The sediment-hosted stratiform copper ore system. Economic Geology One Hundredth Anniversary Volume, 609 to 641. DOI 10.5382/AV100.19. Brown, A.C., 2017. Constraints on conceptual and quantitative modeling of early diagenetic sediment-hosted stratiform copper mineralization. Minerals, 7(10), 192. DOI 10.3390/min7100192. Open access under CC BY. Cox, D.P., Lindsey, D.A., Singer, D.A., Moring, B.C. and Diggles, M.F., 2003. Sediment-hosted copper deposits of the world, deposit models and database. United States Geological Survey Open-File Report 03-107. Figure 11 of this workup is reproduced from its Figure 3. Hayes, T.S., Cox, D.P., Piatak, N.M. and Seal, R.R. II, 2015. Sediment-hosted stratabound copper deposit model. United States Geological Survey Scientific Investigations Report 2010-5070-M, 147 p. DOI 10.3133/sir20105070M. Figure 10 of this workup is reproduced from its Figure 6. Kelepile, T., Bineli Betsi, T., Franchi, F. and Shemang, E., 2020. Partitioning and distribution of silver in sediment-hosted Cu-Ag deposits, evidence from the Ghanzi-Chobe Belt portion of the Kalahari Copper Belt. Ore Geology Reviews, 124, 103663. DOI 10.1016/j.oregeorev.2020.103663. Nghoongoloka, A., Takahashi, R., Manalo, P., Agangi, A., Sato, H., Mocke, H., Nopeia, M., Blamey, N., Enno, T. and Bowell, R., 2024. Genesis of sediment-hosted copper and silver mineralization at the Omatapati prospect, Kaoko Belt, Opuwo district, Kunene region, Namibia. Resource Geology, 74(1), e12341. DOI 10.1111/rge.12341. Figures 6, 8 and 12 of this workup are reproduced from Figures 1, 2b and 10 of that paper. Polish Geological Institute, National Research Institute, copper and silver resources in Poland, pgi.gov.pl, data as at 2015. Squire, R.J. and Keays, R.R., 2024. The role of supermountain belts and climatic controls on the genesis of copper deposits in the Kupferschiefer and the Central African Copperbelt. Mineralium Deposita, 59, published online 9 November 2023, 717 to 732. DOI 10.1007/s00126-023-01227-2. White Metal Resources Corp., news release, 1 October 2021, technical report and mineral resource estimate, Taranis (Okohongo) copper-silver project, Namibia. INV Metals Inc., news release, 17 March 2011, Okohongo deposit drill results, which reports hole INVR-037 and refers to hole 6 from an earlier release that we have not read. 5 THE MACRO AND THE COMMODITY BACKDROP Copper traded at US$6.52 per pound on 2 September 2026, up approximately 43.6% over twelve months. That price is what makes a Namibian exploration story fundable, so it sits behind everything else in this workup. The structural argument for copper is now well rehearsed and largely correct. Electrification, grid rebuilds, electric vehicles and data centre construction are all copper-intensive. Supply is constrained by ageing mines with falling head grades, by a long period of underinvestment in greenfield exploration, and by development lead times the industry itself puts at well over a decade. Kaoko's prospectus cites S&P Global for a lead time averaging 17 years. Kaoko's own market section quotes demand rising from roughly 28 million tonnes in 2025 to 42 million tonnes by 2040, supply peaking near 27 million tonnes around 2030 before declining toward 22 million tonnes by 2040, and a potential structural shortfall of up to 10 million tonnes a year by 2040. Those are third-party numbers from S&P Global, Baker Steel and Fastmarkets. The prospectus discloses, against each of them, that the author has not provided consent for the statement to be included. That is a standard mechanism, and it means the numbers arrive without the author standing behind their use in that document. The macro is best treated as a tailwind rather than as the investment case. A rising copper price lifts every copper explorer's ability to raise money and every copper explorer's share price. It does not make a particular 40 metre interval of malachite economic. If the copper cycle turns, the September placement has bought Kaoko time that most explorers at this stage do not have. The company still has no revenue and no facilities, so a long enough downturn would eventually reach it. Silver and tungsten sit alongside as credits rather than drivers. Silver credits at Chalkos are real, with 2,030g/t reported in surface sampling. Tungsten at Karibib is interesting because Western supply is thin, but at 0.68% tungsten trioxide from a dump sample it is a long way from being a business. 6 HOW KAOKO COMPARES WITH ITS PEERS Section 02 set out what the market is paying for Kaoko today. This section asks a different question. Is roughly A$174 million a normal price for a company at this stage, a high one, or a low one? The only way to answer that is to look at what the market pays for other listed explorers doing similar work on similar ground. Market capitalisation is the share price multiplied by every share on issue. It is what the market says the whole company is worth. For an explorer with no revenue and no mine, it is close to a pure statement of what investors think the ground might one day be worth, discounted for the chance that it never gets there. The peer set, and why these companies Eight companies are compared below. Three are the other ASX-listed copper explorers working in Namibia, which is the closest available comparison on geology and on country. Two work the Kalahari Copper Belt in Botswana, the other large sediment-hosted copper province in southern Africa. Two are ASX-listed copper explorers at a similar stage in South America, included so that jurisdiction is a variable in the table rather than a constant. Kaoko itself is the first row. Jurisdiction, and the evidence each company is standing on Jurisdiction risk is the risk that the country changes the rules. Samso's view is that it deserves more weight than investors usually give it. Attention naturally goes to the grade, the market capitalisation and the newsflow. An explorer has none of those without its licence. An exploration licence is a promise from a government and nothing else until a mine is built, so the things that sit outside an ordinary investment checklist are the things that decide whether an explorer still has a project next year. The table below shows what that means for Kaoko. What the comparison shows Five of the seven peers sit between roughly A$7.7 million and A$24 million. Four of the five are drilling with no Mineral Resource at all. The fifth, Golden Deeps, says it holds resources but does not publish figures we were able to check. That band is where the explorers in this sample have traded through 2026. It is seven companies, not a survey, so it indicates a range rather than establishing one. Kaoko is at roughly A$174 million after the placement. Against that band it is priced at something like seven to twenty-three times its stage peers, on the strength of two holes whose assays have not been returned. Noronex is the sharpest single comparison in the table. It is a copper-silver explorer, in Namibia, drilling, with no company-held resource, and the market values it at about A$7.7 million. That quote had not moved in a session, so the stock is thinly traded and the figure should be read as approximate. Figure 13. Market capitalisation of Kaoko against the seven listed peers in Table 7. The shaded band is the A$7.7 million to A$24 million range in which five of the seven sit, all of them drilling and none with a Mineral Resource it can point to, except Golden Deeps whose claim we could not check. Kaoko is the orange bar, shown at its 3 September 2026 close on all shares on issue after the 7 September placement. The two bars above it are the only companies in the set that have something measured behind them, a JORC Inferred Resource at Midas and a producing mine at Cobre. The figures are not struck on the same day and each is dated in Table 7. Source: Samso Research Media House original figure, built from the figures and sources in Table 7. The two large numbers in the table mark the other end. Midas at roughly A$324 million has a defined 10.5 million tonnes at 1.6% copper. Cobre at roughly A$423 million is producing copper cathode in Chile. Both show what a re-rating looks like when it is attached to something measured. Midas is the nearer case, and its resource was estimated on drilling done before Midas owned the ground rather than on its own campaign. Midas has itself re-rated sharply over the past twelve months. Its shares have traded between 35 cents and A$1.41 in that period, which is the kind of range this cohort moves through in both directions. There is a counterweight. Kaoko holds 39,033 hectares of the Kaoko Belt outright, one half of which carries no recorded exploration of any kind, on a belt with a demonstrated copper-silver endowment set out in Section 04. Scarcity of untested ground in a proven belt is worth something, and part of Kaoko's premium is that. The scarcity argument does not hold as a general claim, and Vantage Metals is why. Vantage began maiden drilling in July 2026 on roughly 3,900 square kilometres of Kalahari Copper Belt ground in Botswana, held outright with no partner. That is about ten times Kaoko's 390 square kilometres, in a belt that already hosts a producing copper mine at Khoemacau. Vantage is capitalised at roughly A$13.5 million. Kaoko is capitalised at roughly thirteen times that. Part of the argument still stands, and it turns on which belt each company is in. Chalkos sits in the Kaoko Belt on the same stratigraphy as Okohongo, which carries 7.7 million tonnes at 1.55% copper and 26.77 grams per tonne silver, and near Omatapati, which is the subject of the published work set out in Section 04. Kaoko's ground is untested but its belt is not. Vantage's Botswanan ground is along strike from a producing mine at Khoemacau, which is the same kind of argument, so this narrows the gap between the two companies rather than closing it. What separates the two companies is a single piece of evidence. Kaoko has copper minerals in core and has told the market about it. Vantage has drilled targets defined by airborne electromagnetic and magnetic surveying. We searched its announcements and its coverage to 3 September 2026 and found no reported intercept, visual or assayed, from that programme. Both are at the same formal stage, which is drilling with no Mineral Resource. Kaoko's market capitalisation is roughly A$161 million above Vantage's. More than A$24 million of that is cash the placement raised, so on enterprise value the gap is closer to A$137 million. That gap rests on the one difference in evidence, together with the belt argument above, before any assay has confirmed either. The assays will settle it. Where this comparison breaks down Three limitations apply, and each one argues for reading the table loosely. The market capitalisations are not struck on the same day, and in a sector this volatile that matters. The oldest figure in the table is Noronex at 23 July 2026 and the newest are dated 3 September. A single announcement can move any of these companies by half in a session, as Kaoko itself demonstrated on 2 September. Five of the seven peers are not pure copper companies. Golden Deeps carries eight metals, Noronex has moved capital into uranium, Mammoth has been expanding into gold, Vantage holds zinc and gold ground in Australia and Argentina alongside its Botswanan copper, and Cobre's market capitalisation is dominated by a producing Chilean mine rather than by its Botswanan exploration. Their market capitalisations are not clean copper-exploration numbers, and Kaoko's is not either, since it includes Karibib. Stage is not a single axis. A company drilling its first hole into untested ground and a company drilling its fiftieth into a known deposit are both described as drilling, and they are not the same proposition. The table separates resource from no resource because that is the one boundary the JORC Code makes objective. Everything on the no-resource side of that line is a matter of judgement. 7 VALUATION, OUR OWN WORKING This is a Track B piece. There is no economic study, no Mineral Resource of any category, and no assay data from the current drilling. Samso puts no price on this company, and would not be able to defend one if we did. What we can do is set out what the market is currently paying for, and what it is comparing the company to. What the market is paying for At the 3 September close of A$2.50 the market is valuing Kaoko at roughly A$174 million on all shares on issue after the placement, against more than A$24 million of cash and two projects with no defined resource. That leaves an enterprise value of roughly A$150 million, on Samso's own arithmetic. The company's September presentation gives its market capitalisation at listing as A$12.1 million, against approximately A$151 million at 3 September before the placement. On the same published figures the market capitalisation has moved by roughly twelve times in four months, and by roughly fourteen times once the placement shares are counted at the same price. It has done so on one announcement containing no assays. That does not make the move wrong. Exploration repricing is often front-run, and by the time assays confirm a discovery the entry price has usually gone. It does mean the risk profile has changed. A buyer in May was paying very little for the geology. A buyer in September is paying mainly for it. The comparable the company itself uses Kaoko's own material points repeatedly at Midas Minerals (ASX: MM1) and its Otavi Project. Midas reported a maiden Inferred Resource of 10.5 million tonnes at 1.6% copper and 21g/t silver, and drill intercepts including 50 metres at 5.55% copper. Kaoko's presentation put Midas's market capitalisation at approximately A$226 million as at 4 May 2026. That figure is now four months old and Midas has re-rated since. Section 06 has it at roughly A$324 million at 3 September 2026, and it is that later number, not the one in Kaoko's slide, that a reader should hold in mind. That comparison is useful, and it needs its counterweight. Midas has a JORC Inferred Mineral Resource. Kaoko has visual logging from two holes. One company has established defined tonnes and grade, and the other has established the presence of visible copper minerals. Most exploration companies never close the distance between those two positions. Framing the upside, with its counterweight If the assays confirm what the core suggests, and if the mineralisation proves continuous across the 2.2km corridor between Otniel and Donkey Hill, then a company at A$174 million with a defined resource in an underexplored belt during a copper bull market could be worth a multiple of that. Midas is the company's own comparable and it is at roughly A$324 million on a 10.5Mt Inferred Resource. Midas got there with a defined resource rather than with visual logging, and its own price has moved a long way in twelve months, so it is a moving benchmark rather than a fixed one. Two things have to be said in the same breath. Midas got there with a defined resource rather than with visual logging, and its own price has moved a long way in twelve months, so it is a moving benchmark rather than a fixed one. The counterweight carries equal weight. Most explorers that log visible mineralisation never convert it into a resource, and visual logging often fails to survive the laboratory. The common sequence is that a geologist logs 6% malachite by volume, the assay returns 0.4% copper, and the share price gives back the move within a day. The shareholders who bought at the top of the visual announcement carry that loss. A market capitalisation that can multiply can also halve. The stock closed at A$0.735 on 1 September and at A$2.50 on 3 September, and the placement was priced at A$2.20. This is a statement about market behaviour and scale. It is explicitly not a valuation, and it is not a reason to act in place of the Samso Call. Newsflow that will establish value Assays from DDOT001 and DDOT002. Expected within four to six weeks of 2 September 2026. This is the settling event for the whole workup. DDOT003, DDOT004 and DDOT005. DDOT003 tests along strike. Until a hole from a second collar hits the same zone, continuity is a hypothesis. The move to Donkey Hill. Donkey Hill is the company's stated primary target and hosts the 69.6% copper grab sample. It has not been drilled. Confirmation of EPL 7608 and EPL 7943 renewals. A dated grant, in an announcement. Karibib geochemistry. A coherent multi-kilometre anomaly would justify the earn-in payments. A weak or incoherent result would raise a fair question about whether they should continue. A second rig. Named in the placement's use of funds on 7 September, and still described in the same week's presentation as under investigation. When it reaches site, and how many metres it adds, is the thing to watch. The next capital raise. Done, on 7 September, at A$2.20 for A$20 million. It was priced 12.0% below the last close and 20.2% above the fifteen-day average. At the June quarter's burn the funding question is settled for some years. 8 WHAT WOULD CHANGE OUR MIND The case getting stronger Assays confirming meaningful copper grades over the logged intervals would establish a discovery rather than a prospect. What counts as meaningful depends on width and depth, and at 40 metres from surface across tens of metres the bar is lower than it would be at 400 metres. Grades of 1% copper or better over broad intervals from that depth would be a serious result. A hole from a second collar hitting the same zone would convert continuity from interpretation to observation. Drilling at Donkey Hill, where the highest surface grades were recorded, intersecting something comparable would suggest a district rather than a prospect. Confirmation of the licence renewals would remove the single largest binary risk under the whole company. One of these tests has already been answered. The capital raise was done on 7 September at A$2.20, a 12.0% discount to the last close and a 20.2% premium to the fifteen-day average, with new and existing long-only global institutions reported on the register. That is a result in the company's favour on funding. The assays are a separate question. The case getting weaker An assay result materially below what the visual logging implies is the most common outcome of this sequence. Copper oxide minerals are highly visible in core, so a small volume percentage can look substantial to the eye. A poor result on true width, where the drilled interval proves to be a shallow angle through a much narrower zone, would shrink the system without changing a single reported number. Failure or delay in the EPL renewals would affect both Chalkos licences at once, and the position remains pending on the public record. Movement on Namibia's draft Minerals Bill toward mandatory local equity would change the value of everything the company holds. The settling event The assays from DDOT001 and DDOT002, expected in the four to six weeks from 2 September 2026. Everything in this workup is provisional until they land. The company has said it is expediting them, which is the right call and also means there is no long window in which to make up your mind. 9 MANAGEMENT AND DNA The board is small, experienced and thinly stretched, which is normal for a company of this size. Gerard O'Donovan is Managing Director and CEO. His background includes senior roles with Pilbara Minerals and Rio Tinto and an executive directorship at Sun Silver. He holds 3,500,000 shares on the current Market Index register and 1.5 million options, and bought 68,490 shares on-market at 48 cents on 22 May 2026. The prospectus recorded 3,750,000 shares at admission. The two figures do not reconcile and the company is where that is settled. On-market buying by a managing director two weeks after listing is a modest but positive signal. Mark Thompson is Non-Executive Chair. He founded Talga Group and led it as Managing Director, and previously worked at Western Mining Corporation and Sons of Gwalia. Talga is a genuine credential. It is a company he took from nothing to a listed graphite developer with a European processing strategy, and it demonstrates a willingness to stay with a story for a long time. Jody Dahrouge is a Non-Executive Director and the technical anchor on the board. He founded Dahrouge Geological Consulting in Canada and specialises in early-stage project generation across base metals, rare earths and uranium. This is the right person to have on the board of a company whose entire proposition is generative exploration. Brett Tucker is CFO and Company Secretary, a chartered accountant with a background in audit, compliance and IPOs. In country, Lisias Pius is Country Manager and Dr Ismael Kangueehi is Exploration Lead, holding a PhD in Earth Sciences from Stellenbosch. Both are Namibian. Kaoko has also engaged local Namibian contractors for site works and drilling, awarding the maiden drilling contract to Optimine Exploration and Drilling Namibia. Callum Standing, a co-founder, was appointed Consulting Geologist in August 2026, having previously worked at SQM, Chalice Mining and Liontown Resources. The DNA read On appetite the record is good. This company listed on 7 May, mobilised field teams within days, reordered its own drill plan in June on the evidence, awarded a drill contract on 23 July, and had two holes in the ground and reported by 2 September. Four months from listing to first drill results is fast, and it was done with two Namibian projects on two different belts. On capability the geological bench is genuine, both on the board and in country. The Namibian presence is operational rather than nominal. Community and conservancy engagement has been reported specifically, and at Chalkos, which is inside the Anabeb Conservancy, that carries more weight than it would in Western Australia. The company's own documents differ on the tenure. The prospectus describes the Chalkos licences as being on State land within the conservancy, and the JORC table in the 2 September announcement describes EPL 7943 as communal land. Two things sit on the other side of the ledger. Directors and management fees are budgeted at A$518,840 a year against a two-year Karibib exploration budget of A$500,000. That ratio is not unusual for a small ASX explorer and it is still worth stating plainly. And there is no long shared track record here. Kaoko is fourteen months old as a corporate entity. The individuals have history. The team, in this configuration, does not. The board raised A$20 million within a week of the drilling announcement, priced 12.0% below the last close and 20.2% above the fifteen-day average, and put the money against a second rig and an expanded programme. On appetite, and on the ability to raise, that is a good result. It says nothing about what the assays will show. 10 SAMSO CONCLUDING COMMENTS SAMSO CALL: WATCH & DYOR The settling event we are watching is the assay results from DDOT001 and DDOT002, expected within four to six weeks of 2 September 2026. I have watched this exact sequence of events play out many times on the ASX. A junior explorer drills its first holes, the geologists see copper in the core, the company writes a careful announcement with cautionary notes in bold on every page, and the market ignores every one of those notes and doubles the stock in a session. Sometimes the assays arrive and confirm it. More often they do not, and the same market that doubled it takes it all back in a morning. What makes Kaoko worth the attention is not the price move. It is the ground. Two exploration licences covering 39,000 hectares, one of which has never had a single recorded exploration programme run over it, in a belt that is geologically similar to the Central African Copperbelt and that nobody has systematically tested with modern methods. That is a genuinely scarce position in 2026. Most of the world's easy copper ground was picked over decades ago. The company's own independent geologist is straight about the limits of the analogy. "Whilst the Kaoko Belt shares similarities with the Central African Copperbelt, it is not an exact analogue." "The Kaoko Belt is Neoproterozoic in age and may suggest a different style of mineralisation which will be expanded on with further exploration and understanding." Kaoko Metals Prospectus, 23 February 2026, Section 4.5, summarising the Independent Technical Assessment Report The Zambian and Congolese systems are copper-cobalt and the Kaoko Belt is copper-silver. Most economic sediment-hosted copper occurs in basins older or younger than this one. Those caveats sit in the prospectus in the geologist's own words, and I would rather see them there than not. Sixty metres of visible copper from 36 metres down, on the first attempt, on ground nobody has drilled before, is a real result. It is also still only a photograph. The distance between what the eye sees and what the laboratory measures is where this investment sits at the moment, and that distance closes within weeks. The A$20 million raise answers a question that hangs over every explorer at this stage, which is whether the company can fund the work in front of it. Kaoko now has over A$24 million of cash, so it can. It will need a great deal more drilling to prove this up, and drilling is expensive. Two holes and a set of visual logs are the beginning of the work rather than the end of it. The raise lets the company keep a rig turning and keep the newsflow going while it finds out what it has. It also means the next raise is not priced off whatever the assays say. WA1 Resources is the comparison I keep coming back to. WA1 drilled its first programme in the West Arunta in Western Australia in November 2022 and made the Luni discovery, which the company describes as the most significant niobium discovery in more than seventy years. Kaoko could turn out to be a similar situation, a fresh discovery on ground nobody has tested that develops into a Tier 1 deposit. Tier 1 is industry shorthand for a deposit big enough, rich enough and long enough lived that the major mining companies want to own it. I do not know whether that is what this is, and neither does anyone else at this point. The part worth taking from the comparison is how long the road is even when it works. WA1 made its discovery in November 2022 and published its Mineral Resource in May 2026, and the pre-feasibility study is still not finished. That is three and a half years of drilling after the discovery hole. Kaoko is at the beginning of that road and the assays are the first step along it. Source. WA1 Resources project page, Luni Mineral Resource of 220 million tonnes at 1.0% Nb2O5 dated 13 May 2026, and the company's own account of the discovery. So the call is WATCH & DYOR, and I want to be specific about what the DYOR means here, because it is not a hedge. It means do the work now, in the next four weeks, while the assays are still in the laboratory. Read the 2 September announcement in full, including Appendix A and the JORC table. Understand that the visual estimates are volume percentages and not grades. Look at the escrow schedule and know when the 11.57 million shares come free. Ask the company about the licence renewals and about the vesting conditions on the 5.55 million performance rights. Decide, in advance, what assay result would make you a buyer and what result would make you walk away. Then wait for the number. What I would not do is decide anything about this company on the basis of a core photograph and a 151% day. That is not because the photograph is misleading. It is not, and I think the geology behind it is interesting. It is because the price already assumes an answer to a question that has not been asked yet. So the question worth sitting with is this. Are you willing to pay today's price for an answer that has not arrived yet? 11 DISCLAIMER The information contained on this website is the writer's personal opinion and is provided to you for information only and is not intended to or nor will it create/induce the creation of any binding legal relations. There may be a conflict of interest present with commercial arrangements with companies and/or stock held. Samso or an associate may receive a commission for funds raised. The information or opinions provided herein do not constitute investment advice, an offer or solicitation to subscribe for, purchase or sell the investment product(s) mentioned herein. It does not take into consideration, nor have any regard to your specific investment objectives, financial situation, risk profile, tax position and particular, or unique needs and constraints. Accordingly, no warranty whatsoever is given, and no liability whatsoever is accepted for any loss arising, whether directly or indirectly as a result of this information. Investments are subject to investment risks, including possible loss of the principal amount invested. The value of the product and the income from them may fall as well as rise. You may wish to seek advice from an independent financial adviser before committing to purchase or invest in the investment product(s) mentioned herein. If you choose not to do so, you should consider whether the investment product(s) mentioned herein are suitable for you. The writer will not, in any event, be liable to you for any direct/indirect or any other damages of any kind arising from or in connection with your reliance on any information in and/or materials appended herein. The information and/or materials are provided "as is" without warranty of any kind, either express or implied. In particular, no warranty regarding accuracy or fitness for a purpose is given in connection with such information and materials. The Samso Way – Seek the Research Here at Samso, we pride ourselves on delivering content for investors that is independent and informed by over three decades of experience in the industry. Our content is well-researched and is only created if I see merit in discussing the company's story. Our mission is simple. Cut through the noise and spotlight what matters. Genuine stories, grounded insights, and real opportunity. Investors can explore our four core platforms: Coffee with Samso Samso Insights Samso News Samso Research There may be numerous paths to success in investing, but the common thread among successful individuals is that they remain committed to making informed decisions. Equip yourself with the right knowledge and tools, and you will be well on your way to achieving your financial goals. The Samso Philosophy: Stay curious. Stay sharp. And remember, digging deeper always uncovers the real value. Happy Investing, and the only four-letter word you need to know is DYOR. Disclaimer This Samso Research Workup is independent research commentary for general information only. Nothing in this document is financial product advice, and it does not take into account any reader's objectives, financial situation or needs. Figures are drawn from public sources believed reliable at the stated dates but are not guaranteed; market-sensitive numbers change daily. Samso or associated parties may hold positions in, or have commercial arrangements with, companies mentioned. Seek professional advice before making investment decisions. Read full Disclaimer. About Samso Samso is a trusted platform that equips dedicated investors with up-to-date industry knowledge and insights from top CEOs and thought leaders. Samso Research | www.samso.com.au | An Investor Lens on ASX-Listed Companies

Updated 6 October 2026. This workup was first released in three parts on 24 September, 29 September and 1 October 2026, and it has been updated since that first release. The update takes in Brightstar's Mineral Resource statement of 28 September 2026, its Sandstone metallurgy results of 24 September 2026 and its drilling announcements to 5 October 2026. Share prices, the peer table and the gold price are now as at 5 October 2026. Brightstar is now labelled a near-term producer. Samso uses that label for a company with a published feasibility study and a stated date for first gold. Contents 11 The Disclosure Record 15 Samso Concluding Comments 12 The Board, Management and Incentives 16 References and Sources 13 Where Brightstar Sits Against Its Peers 17 Disclaimer 14 The Risks Specific to Brightstar 11 The Disclosure Record On 14 April 2026 Brightstar released a Clarification Statement covering its investor presentation of 18 March 2026. The company's own words are the clearest summary. "It has come to the Company's attention that certain statements made in the Presentation may constitute production targets." Four things were withdrawn or retracted. The first was the Sandstone processing plant throughput range of 4 to 5 million tonnes a year. The second was Sandstone production shown in the group production growth chart. The third was an illustrative table of mill throughput against head grade. The fourth was a set of three growth targets. The first of those was a long-term Mineral Resource growth target of 3.0 to 4.0 million ounces. The second was a maiden Ore Reserve target of more than 1.0 million ounces supporting a 6 to 8 year mine life at the Pre-Feasibility Study. The third was an Ore Reserve of 1.5 to 2.5 million ounces supporting a mine life beyond 10 years at the Definitive Feasibility Study. The company's reasoning is set out in the announcement. "It has come to the Company's attention that given the amount of work undertaken on the Sandstone project, in particular, the anticipated completion of the PFS this year and the sizeable existing Mineral Resource estimate of 2.4Moz @ 1.5g/t Au, the statements in relation to targeted Mineral Resource and Ore Reserves growth are too specific and potentially constitute predictions or potential targets from Brightstar about probable future outcomes, as opposed to purely aspirational statements. Accordingly, Brightstar withdraws its disclosure of its targeted growth in Mineral Resources and Ore Reserves until such time as it has completed the relevant studies and has sufficient reasonable grounds to support forecast information in relation to production from Sandstone, and shareholders must not rely on this information." Brightstar Resources, "Clarification of production targets and aspirational statements", ASX announcement, 14 April 2026[R29]

Part 1 mapped a register of tin and tungsten deposits that have been known for a century. This note explains the variable that decided which of them became mines: the size of the ore mineral grains. Grain size sets recovery, recovery multiplies grade, and no drill intersection reports it. Eastern Australia hosts all four answers to the problem, side by side. Samso Insights Research Insights The Eastern Australian Tin - Tungsten Register - part 2 of 3 Samso Geology Series WHAT THIS PIECE COVERS 1.00 THE PREMISE Why every deposit has two grades, and why only one of them is assayed. 2.00 THE MAP Where the four archetypes sit, redrawn from Part 1. 3.00 THE PHYSICS The gravity window, in plain English. 4.00 THE PUBLISHED PROOF Taronga against Windermere, and what that pairing demonstrates. 5.00 ARCHETYPE A Coarse veins, the ore that built the district. 6.00 ARCHETYPE B Locked skarns, tin you can assay but cannot recover. 7.00 ARCHETYPE C Watershed, the fine-grained skarn with a 2026 answer. 8.00 ARCHETYPE D Heemskirk, and what clean ore looks like. 9.00 THE MODERN ANSWERS Ore sorting, fuming, and the strongest case against this note. 10.00 THE LEDGER Effective recovered grade, and how it re-sorts the whole register. 1.00 — THE PREMISE Every deposit has two grades Part 1 closed with three exhibits. Taronga, at 0.10% tin, has a completed definitive feasibility study. Windermere, at 0.27% tin, nearly three times the grade, has never been mined through any price cycle since its discovery. Gilmore's ore ran 7.6% tin and paid for itself with 1930s technology, while Gillian's 0.65% defeated a 2013 study advised by the world's largest tin producer. If grade decided development, every one of those outcomes would be reversed.[1][6][8][9][11] The variable that reconciles them is grain size. A tin deposit really has two grades: the assay grade a laboratory reports, and that grade multiplied by the recovery a processing plant can actually achieve. Grain size sets the second one, because every recovery machine ever built has a particle-size floor, and mineral grains finer than the floor leave the plant in the tailings no matter what the drill intersection said. The result is an invisible cut-off grade that never appears in a resource table.[2] This matters more in 2026 than it has in forty years, because the tin price is at records and the temptation is to read every old intercept at today's prices. Price is a multiplier on recovered metal, not on assayed metal. A higher tin price can pay for more grinding, more flotation and more furnace time; it cannot make a cassiterite grain bigger. This note works through the physics in plain English, then walks the four metallurgical archetypes that Eastern Australia hosts within two belts.

Abstract Tusker Minerals (ASX: TSK) has put a 2.1 to 2.6 billion tonne JORC Exploration Target in front of the market on the Douala Basin in coastal Cameroon, containing an indicative 7 to 8 million tonnes of rutile within roughly 40 kilometres of a deep water port. It is a genuinely large number on ground that has barely been drilled. It is also conceptual by definition, and the company says so. Two things need correcting before anyone reads the grade. First, a figure of 0.03 to 0.05 per cent rutile is circulating in the public domain and it is wrong by a factor of ten. The announced grade is 0.3 to 0.35 per cent, and we show the evidence for that in Section 3. Second, 70 per cent of the heavy mineral content is kyanite, which the company assigns no value and has not tested. Strip it out and the 2.1 to 2.3 per cent heavy mineral headline is closer to about 0.6 per cent of minerals anyone currently pays for. On our own comparison, the modelled grade is about a third of Sovereign Metals' Kasiya and sits at or barely above the cut off grade Sierra Rutile used in its own audited resource statements. What that means for economics is not something anyone can settle yet. The company has no mineral resource on any asset, no metallurgical testwork anywhere in the portfolio, and its flagship Cameroon tenure is still under application rather than granted. The reason to keep watching is narrow and real. Seven reconnaissance auger holes returned 0.33 to 0.76 per cent rutile, two to two and a half times the modelled target grade, and they sit outside the target footprint. Either the target understates the deposit or those holes are the good bits. Nobody knows yet. Funding is the near term issue. A$1.35 million of cash at 30 June 2026 against roughly A$5.1 million of annual all in spending, with the gap meant to be closed by the sale of the Malawi subsidiary to AuKing Mining for up to A$4.85 million. That transaction is binding but not complete, with a long stop date of 30 September 2026. The last equity raise was at 30 cents. The shares are at 7.3 cents. This company's re-rating has already happened. DY6 rose about 475 per cent through 2025 and roughly 78 per cent of that has been given back. Our call is Watch and DYOR, and the event we are watching is the maiden JORC Mineral Resource at Douala Basin. Note the company's six to nine month guidance covers the exploration work that tests the target, not the resource itself, and it says no assurance can be given that a resource will be defined. Contents 1 Why We Are Doing a Full Workup on This One 6 What Would Change Our Mind 2 The Company and The Story 7 Management and the DNA Question 3 The Assets 8 Samso Concluding Comments 4 The Commodity Backdrop 9 Disclaimer 5 Where This Sits Against Its Peers Tusker Minerals Snapshot This is a living document and the share price will have moved. All market data is as at 27 August 2026 unless stated otherwise. Shares on issue and the option and performance right counts are taken from the company's ASX filings of 20 and 27 August 2026. Metric Value Share price A$0.073 (ASX, 27 August 2026) Market capitalisation About A$8.16 million Shares on issue 111,759,425 (ASX filing, 27 August 2026) Cash A$1.35 million at 30 June 2026 (Appendix 5B) 52 week range A$0.062 to A$0.345 One year return About negative 76 per cent Options and performance rights 66,096,796 options and 23,675,000 performance rights, about 1.8 times the share register Last equity raise September 2025, at A$0.30 JORC Mineral Resources None on any asset Broker coverage None identified Substantial holders, from the last lodged notices, all of which pre-date the August 2026 issue of new shares. Orwellian Investments and Bridge The Gap Trading Pty Ltd, 12.40 per cent, 22 September 2025. Elias Pungong Prombo Pupesie, 7.96 per cent, 30 December 2025. Zhenshi Group (HK) Heshi Composite Materials Co Limited, 7.67 per cent, 22 September 2025. Matthew Vincent Horgan, 6.68 per cent, 30 December 2025. Zhenshi Group was a A$1.5 million cornerstone investor in the 2023 IPO. Share price context. This is a stock that has already run and already given it back. DY6 Metals rose roughly 475 per cent through 2025 on the Cameroon rutile acquisition, reaching a market capitalisation of about A$17 million. It listed in June 2023 at 20 cents, raised at 30 cents twice in 2025, and last traded at 7.3 cents. The 52 week high of A$0.345 and the low of A$0.062 both sit inside the last twelve months. A five bagger from here would be the second one, after the first was given back. 1 Why We Are Doing a Full Workup on This One I have been following this company since it was DY6 Metals. Samso has published on it eight times in fourteen months, which is one of the longest continuous threads we have on a single stock. I sat down with Cliff Fitzhenry for Coffee with Samso in September 2025, when the Cameroon rutile story was new and almost nobody on the ASX was looking at it. In October 2025 I wrote that a genuine discovery at Tundulu could put this company somewhere near A$100 million. It did not happen. The shares went from about A$15 million of market capitalisation in July 2025, up hard through the second half of that year, and back down to about A$8 million today. Tundulu has since been sold. This is where the Samso research workup comes into play. When you have been on a story through a full round trip, you either walk away from it or you go back and do the deepest piece of work you know how to do. This company has changed what it is twice in three years. It has just sold assets to add non-dilutionary funds for the existing projects. In addition, the recent news from management has put a big number for its valuation on the ground, which has not had modern systematic exploration activities. So the question this workup answers is not whether the Douala Basin is big. The company has told us it is big. The question is whether what is in it is worth anything, and whether this team can prove it before the money runs out. 2 The Company and The Story Tusker Minerals Limited is a Perth registered explorer with rutile and heavy mineral sands ground in Cameroon, and one retained rutile project in northern Malawi. It listed on 29 June 2023 as DY6 Metals Limited after a A$7 million IPO at 20 cents. Shareholders approved the name change at the November 2025 annual general meeting, and the ticker moved from DY6 to TSK on 1 December 2025. The words you need before we go further Rutile is the highest grade natural form of titanium dioxide, at 95 per cent TiO2 or better. Titanium is not sold as ore. It is sold as feedstock, graded by how much titanium dioxide it carries, and rutile sits at the top of that ladder because nature has already done the upgrading. Ilmenite, at 30 to 60 per cent TiO2, is the cheap and abundant workhorse and makes up the great bulk of the market. Heavy mineral sands, or HMS, is the deposit style. Heavy minerals are the dense fraction of a sand or a weathered soil. The important thing, and it matters a great deal in this piece, is that only some of those heavy minerals are worth money. Saprolite is deeply weathered rock that sits above fresh bedrock. It is soft. That is why this style of deposit can often be dug without blasting, which is the whole cost argument behind projects like Sovereign Metals' Kasiya in Malawi. An Exploration Target is not a Mineral Resource. Under the JORC Code it is a conceptual estimate of what might be there. The Code requires the company to state that there has been insufficient exploration to estimate a Mineral Resource, and that it is uncertain whether further exploration will result in one. Tusker has stated exactly that. Every number in this workup that comes from the Douala Basin target carries that qualification. The strategy as it actually is The stated strategy is titanium feedstock in Cameroon, funded by selling the Malawi rare earths portfolio. That is a fair description of where the company is pointed today. It is not a complete description of how it got there. The record reads like this. It listed in June 2023 as a Malawi heavy rare earths and niobium explorer with Machinga as the flagship. In August 2023 it hired Lloyd Kaiser as chief executive, a genuine development hire with ten years at Arafura Rare Earths in offtake and engineering roles on Nolans Bore. In April 2025 it acquired two Cameroon rutile and heavy mineral sands companies and the story became titanium. In July 2025 Kaiser left and Cliff Fitzhenry, an exploration geologist, became chief executive. In January 2026 the company announced a strategic review to unlock value from its Malawi rare earth and gallium assets. Between April and August 2026 it sold Tundulu, then agreed to sell Machinga, then restructured that into the sale of the entire Malawian subsidiary. The review has now been executed almost in full. What emerges is a Cameroon rutile explorer with a Malawi remnant. Whether a reader calls that focus or drift is a fair question, and we come back to it in Section 8. Capital structure The share count is small for a company that has been issuing paper, and there has been no consolidation. The April 2026 bonus loyalty option issue was one option for every seven shares held. Five million shares were then issued on 19 August 2026 as part consideration for Yaounde West, on conversion of vendor performance rights, with a further 2,142 shares issued at the same time. The figure above is taken from the company's own cessation notices lodged on 27 August 2026 rather than from a vendor data feed. Several widely used market data services still carry 106.76 million shares and a market capitalisation near A$7.8 million, because they have not picked up the August issue. Options and rights This is the part of the capital structure that is easy to miss, and it is larger than the share register. So against 111,759,425 shares on issue there are 66,096,796 options and 23,675,000 performance rights. On a fully diluted basis that is roughly 201.5 million securities, close to 1.8 times the current share register. Most of it is a long way out of the money at 7.3 cents. The 21,900,000 TSKAP options struck at 60 cents would need an eightfold move before they matter. But 26,409,362 TSKAN options are struck at 8 cents and run to January 2029, which is roughly a tenth of a cent above the current price. Those are, for practical purposes, at the money. We are not going to tell a reader what that means for the share price, because we do not know. What we will say is that anyone modelling a re-rating here should model it on a share count well above 111.8 million, and that the performance rights in particular need to be understood before the dilution is treated as theoretical. Funding, stated honestly There are two defensible readings of the runway and they are far apart. On the Appendix 5B convention, which counts operating outflow only, A$1.35 million divided by A$263,000 gives about five quarters. That reading flatters the company, because exploration spending sits in the investing line for Tusker and is therefore excluded from the calculation. Including exploration, all in spending across the twelve months to June 2026 was roughly A$5.1 million. On that basis A$1.35 million is about one quarter of activity. The gap is meant to be closed by asset sale proceeds rather than a placement. That is management's own framing, described as non dilutive funding. It is worth noting that there has been no cash raise since September 2025, eleven months, and that the last two raises were both done at 30 cents against a 7.3 cent share price today. A raise from here would be heavily dilutive, which is precisely why the asset sales matter as much as they do. The Malawi divestment Two separate transactions, same buyer, AuKing Mining (ASX: AKN). Tundulu, announced 17 April 2026, up to A$5.55 million. A$50,000 non refundable deposit for 40 day exclusivity, A$1.0 million cash at completion, A$2.2 million in staged deferred cash, A$1.25 million in AuKing shares and A$1.0 million in performance rights tied to resource milestones. AuKing funded it with a A$3 million raise. This transaction has completed and AuKing is drilling the project. Tusker shares rose 43.8 per cent on the day of announcement. Green Exploration Limited, announced 18 August 2026, up to A$4.85 million. This is a binding share sale agreement over 100 per cent of Tusker's Malawian subsidiary, which holds Machinga, Ngala Hill, Salambidwe and Karonga. It replaces an earlier agreement of 3 August 2026 over Machinga alone, which was for up to A$4 million. Anyone still quoting A$4 million is quoting a superseded deal. The Mzimba rutile licences transfer out of Green Exploration before completion and are retained. Conditions precedent include regulatory, governmental and shareholder approvals, AuKing shareholder approval for the scrip component, and ASX confirmation that the acquisition does not require re-compliance with listing rule Chapters 1, 2 and 11. Malawi’s mining regulator must approve the licence transfers. The long stop date is 30 September 2026 and the transaction had not completed as at the date of this workup. That is a month away and it is the near term thing to watch. Across both deals the headline consideration is up to A$10.4 million against a market capitalisation of about A$8.2 million. Before anyone gets excited about that comparison, most of it is deferred, in scrip, or contingent on milestones that depend on AuKing’s own drilling success rather than anything Tusker controls. 3 The Assets A correction to the public record. Before we discuss grade, a number in circulation needs fixing. A reader who searches for the Douala Basin rutile grade will find 0.03 to 0.05 per cent. That figure is wrong by a factor of ten. It appears to be a decimal point error in one trade publication and it has been repeated from there. The grade Tusker announced on 14 July 2026, and carried unchanged into the amended announcement of 17 July 2026, is 0.3 to 0.35 per cent rutile. We are not asking anyone to take our word for it. The evidence is set out below. Four sources agree. One does not. The arithmetic inside the announcement settles which is right, and it does so twice. Cross check one. The announcement reports 7 to 8 million tonnes of contained rutile inside 2.1 to 2.6 billion tonnes of material. Divide one by the other and you get 0.27 to 0.38 per cent rutile. At 0.03 to 0.05 per cent, that same tonnage would contain 0.6 to 1.3 million tonnes of rutile, not 7 to 8 million tonnes. Cross check two. The announcement reports total heavy minerals of 2.1 to 2.3 per cent, of which rutile is 15 per cent of the mineral assemblage. Multiply those together and you get 0.32 to 0.35 per cent rutile. That matches the announced figure. It does not match 0.03 per cent. Both checks land in the same place. This cuts in the company's favour, and we want to be even handed about saying so. If you were carrying the wrong number, the deposit is ten times better than you thought it was. It also means the grade is high enough to be worth arguing about properly, which is what the rest of this section does. A tenth of a per cent would not have been worth the argument. 3.1 Douala Basin heavy mineral sands project, Cameroon Figure 1. Map of Cameroon showing Tusker Minerals Central Rutile Project in green, and the Douala Basin Project in blue, encompassing +11,000km2 of highly prospective tenure. (Reproduced from Tusker Minerals, Amended Announcement, Douala Basin Exploration Target, ASX, 17 July 2026. The figure is unchanged from the 14 July original. Caption is the company’s own.) This is the reason for the workup. Tusker holds 100 per cent through the acquisition of Rhino Resources Ltd, announced 25 April 2025. The tenure is 2,580 square kilometres of coastal ground, three granted exploration licences plus three valid applications, and it incorporates what was previously called the Missole Project. On infrastructure, the announcement states the project is located within approximately 40 kilometres of the Port of Douala, and its map draws indicative logistics catchment zones at 25 kilometres and 50 kilometres. Those are the figures to use. That is genuinely good, and better than most large African mineral sands projects. Figure 2. Douala Basin Project tenure relative to the Port of Douala, showing indicative 25 km and 50 km logistics catchment zones and the Project’s strategic proximity to critical export infrastructure. (Reproduced from Tusker Minerals, Amended Announcement, Douala Basin Exploration Target, ASX, 17 July 2026. The figure is unchanged from the 14 July original. Caption is the company’s own.) Figure 3. Diwong South Exploration Target showing modelled THM distribution on the left and rutile as a percentage of THM on the right. (Reproduced from Tusker Minerals, Amended Announcement, Douala Basin Exploration Target, ASX, 17 July 2026. The figure is unchanged from the 14 July original. Caption is the company’s own.) Figure 4. Diwong South Exploration Target extent within Tusker’s broader project tenure, highlighting the footprint of the current multi-billion-tonne target. (Reproduced from Tusker Minerals, Amended Announcement, Douala Basin Exploration Target, ASX, 17 July 2026. The figure is unchanged from the 14 July original. Caption is the company’s own.) The potential quantity and grade of the Exploration Target reported in this announcement is conceptual in nature. There has been insufficient exploration to estimate a Mineral Resource, and it is uncertain if further exploration will result in the estimation of a Mineral Resource. That statement is the company's own, and it is mandatory under the JORC Code. It is reproduced here because it tends to get lost under the word “billion”, and it is the sentence that tells a reader what an Exploration Target is. The announcement is also specific about what the estimate is missing, and this is worth quoting because companies do not always spell it out. The stated limitations are incomplete historic assay support in parts of the dataset, the absence of project specific density data, conceptual vertical model constraints, and limited metallurgical data. Density matters more than it sounds. Without it, a tonnage figure rests on an assumed conversion from volume to mass. 3.2 What is actually in the rock The 2.1 to 2.3 per cent heavy mineral grade is not 2.1 to 2.3 per cent of saleable product, and the difference is the whole of this section. Figure 5. What is in the Douala Basin heavy mineral fraction. Kyanite is seventy per cent of it, and the company has assigned it no value pending testwork and market assessment. (Samso Research Media House.) Seventy per cent of the heavy mineral content is kyanite. This is Samso analysis of what that means, and we want to be clear it is ours, built from the company’s own disclosed numbers rather than handed to us. Kyanite is an aluminium silicate used mainly in refractories, which are the heat resistant linings inside furnaces. It is a real mineral with a real market. It is not, by the mineral sands industry’s own definition, a valuable heavy mineral. The standard industry primer, Greg Jones's "Mineral Sands: An Overview of the Industry" written for the AusIMM, defines the valuable heavy minerals as ilmenite, leucoxene, rutile and zircon, and classifies kyanite, staurolite and garnet outside that group. Jones wrote it while he was at Iluka Resources, and he is the Competent Person on this exploration target. We are not arguing against the company’s expert. We are quoting him. The company itself assigns kyanite no value, and it is careful about this. The announcement references kyanite product specifications from Kyanite Mining Corporation and the USGS Mineral Commodity Summaries 2026, then states plainly that no assumption is made that Douala material would achieve comparable product specifications, recoveries or pricing. It says recovery, product quality and commercial potential all remain subject to further testwork and market assessment. That is a fair disclosure and we are not arguing with it. The market size is the harder constraint. Global production of the whole kyanite, andalusite and sillimanite group runs at roughly 470,000 tonnes a year, at a United States price of about US$420 a tonne. Tusker’s exploration target implies something in the order of 34 to 39 million tonnes of contained kyanite. Even a modest operation would swamp that market many times over, and this is a bulk, low value, high freight product that would need to be shipped out of West Africa. So the arithmetic that follows is ours. If the kyanite turns out not to be saleable, the 2.1 to 2.3 per cent heavy mineral headline is closer to about 0.6 per cent of heavy minerals that anyone currently pays for, being rutile at 15 per cent, zircon at 3 per cent and ilmenite at 9 per cent of the assemblage, which is 27 per cent of 2.1 to 2.3, or 0.57 to 0.62 per cent. On that scenario the kyanite is process load rather than product, mass that has to be mined, hauled and separated for no revenue. We are not asserting that outcome. We are saying it is the scenario a reader should hold in mind until testwork says otherwise, because nothing published so far rules it in or out. Testwork has not been done. A kyanite product specification and a named buyer would change the picture materially, and it is one of the things we are watching for. The narrow point we will make is this. Nobody should read the headline heavy mineral grade as though the whole of it is currently saleable, because on the company’s own disclosure that has not been established yet. 3.3 The drilling, and the part of the story that is genuinely interesting Figure 6. Diwong South Exploration Target showing selected historical sonic and hand-auger intercepts together with Tusker reconnaissance auger results. (Reproduced from Tusker Minerals, Amended Announcement, Douala Basin Exploration Target, ASX, 17 July 2026. The figure is unchanged from the 14 July original. Caption is the company’s own.) Two observations matter here, and they pull in opposite directions. The first is that these holes are running well above the modelled target grade. The best of them is more than twice it. They also sit outside the Diwong South target footprint. The chief executive has made this point directly. In the 17 July 2026 announcement Cliff Fitzhenry said, "The Douala Basin Exploration Target is a game-changing milestone for Tusker, establishing genuine multi-billion-tonne scale within a shallow, infrastructure-advantaged rutile-zircon system." The announcement then says the target covers only a portion of one licence while reconnaissance drilling is already identifying materially higher grade near surface zones. The scale claim is fair on the company's own numbers. What the quote does not say, and could not yet, is anything about grade continuity, because seven auger holes cannot establish it. If the reconnaissance grades hold up over area, the target as modelled understates what is there, and the bull case rests almost entirely on this. The second is that seven hand auger holes across 2,580 square kilometres is not a data set. A 75 millimetre auger to a few metres depth is the cheapest reconnaissance tool there is. It is the right tool for this stage and we are not criticising its use. It simply cannot tell you whether a grade is continuous, and continuity is the whole question. Historical drilling by Eramet, the French mining group, gives some comfort on thickness. Sixty sonic holes for 1,080 metres and 39 hand auger holes returned intervals including 10 metres at 5.0 per cent total heavy minerals from surface, 12 metres at 4.7 per cent from surface, and 18 metres at 3.7 per cent from surface ending in mineralisation. Those are heavy mineral grades, not rutile grades. Historical drilling bottomed out at roughly 18 to 24 metres. There is no published metallurgical testwork on the Douala Basin. No recovery figures, no product specification, no payability work. For a heavy mineral sands project that is a significant hole, because recovery and product quality are what turn a grade into revenue. The company has built its own heavy mineral sands laboratory in Yaounde, which is a sensible and unusually practical step for a company this size. On timing, read the company’s words carefully, because the amended announcement of 17 July 2026 is more careful than the original. It sets out a programme of exploration activities designed to test the exploration target, being infill and extensional drilling, density and metallurgical work, and updated geological and block modelling. Those activities are targeted for completion within approximately six to nine months, subject to drilling outcomes, permitting and assay turnaround. The company then says that successful outcomes would provide a stronger basis for estimating a maiden resource, and that no assurance can be given that a Mineral Resource will be defined, or that any part of the exploration target will be converted into one. Six to nine months is the timeline for the work, not a timeline for a resource. Anyone reading it as the latter is reading something the company did not say. 3.4 Central Rutile Project, Cameroon The nominal flagship by area, and the one with the deepest history. The area figures in the company’s own materials range from 2,140 square kilometres in April 2025 to 5,901 in July 2025 to about 8,780 on the current project page and about 7,500 on the homepage. We use approximately 7,500 square kilometres across 21 exploration permits and note the inconsistency rather than pretending it is not there. All of those permits are applications, not grants. That is a material point and it belongs in any assessment of this company. The geology is residual and eluvial rutile hosted in saprolite over kyanite bearing mica schist and garnet paragneiss. Graphite appears with depth as oxidation decreases, which is the same association Sovereign Metals reports at Kasiya. The ground borders Lion Rock Minerals’ Minta project. The area produced 15,000 tonnes of high purity rutile between 1935 and 1955, when Cameroon was the world’s third largest rutile producer. Reported assays from the September 2025 programme include 8.7 per cent heavy minerals including 0.6 per cent rutile over the first 2.25 metres, 5.29 per cent heavy minerals including 0.78 per cent rutile over the first 3.5 metres, and individual occurrences at 1.57 and 2.1 per cent rutile. There is no exploration target and no mineral resource. Assays from the most recent programme were still outstanding as at 20 August 2026. 3.5 Mzimba, Malawi, and Yaounde West, Cameroon Mzimba is the only Malawi ground Tusker keeps. A 710 square kilometre licence package in northern Malawi, of which about 50 square kilometres has been tested. Forty one soil and rock chip samples returned a peak of 1.88 per cent TiO2, with 11 of 41 samples above 1 per cent. X ray diffraction confirmed the titanium occurs predominantly as rutile with minor anatase and no detectable ilmenite, which is the commercially preferable outcome. One qualification is essential. These are percentages of titanium dioxide in surface soil and rock chip samples. They are not drill intercepts and they are not in situ rutile grades over a mining width. They cannot be laid alongside Douala’s 0.3 to 0.35 per cent or Kasiya’s 1 per cent without saying so, and any comparison that does lay them side by side is misleading. There has been no drilling at Mzimba. Yaounde West is seven exploration permits acquired in September 2025 for A$1.5 million plus shares. Its attraction is that it covers the only historically recorded area of residual rutile mining in Cameroon. There is no drilling, no assay, no target and no resource. Five million vendor performance rights remain outstanding, contingent on the licences being granted, and no grant has been announced. 4 The Commodity Backdrop 4.1 Why rutile is a different market from titanium Two end markets sit on top of that ladder and they behave nothing like each other. Pigment is about 95 per cent of demand, the white in paint, plastics and paper, and it is a housing and construction cycle business rather than a growth story. Titanium metal is about 5 per cent, going into aerospace, defence and medical use, and it is where the strategic story lives. One more distinction decides whether rutile demand actually grows. The chloride pigment process, used by Western producers, needs high grade feedstock. The sulphate process, mostly Chinese, runs on cheap ilmenite. High grade feedstock demand therefore tracks chloride pigment share, not total pigment demand. 4.2 The supply argument, which is real This is the table that matters. World rutile is a market of roughly 450,000 tonnes a year. Ilmenite is 9.4 million tonnes. Four countries make up about 91 per cent of rutile supply. It is a small, concentrated market, and small concentrated markets move hard when something goes wrong in them. Several things have gone wrong at once. Base Resources’ Kwale mine in Kenya closed in December 2024 on reserve depletion, which is why Kenya drops from 41,000 tonnes to nothing. Iluka produced 10,600 tonnes of rutile in the June 2026 quarter, down from 29,900 tonnes in the December 2025 quarter, and has kept both of its synthetic rutile kilns idle. Rio Tinto approved US$473 million for Zulti South at Richards Bay in March 2026 but was explicit that the project is not about expansion, since it replaces declining production from Zulti North. And Sierra Rutile, the world’s largest natural rutile mine, was taken private by Leonoil and delisted from the ASX on 3 October 2024. There has been no public operational disclosure since September 2024. The single largest source of natural rutile on earth is now an opaque private asset with no reporting obligation. Anyone modelling this market is working from USGS estimates, not from disclosure. 4.3 What the prices actually say The price data does not support the promotional version of this story. Iluka’s realised rutile price was US$1,254 a tonne across the first half of 2026, against US$1,216 a tonne for the 2025 financial year. That is broadly flat. There is no price spike in the data. Ilmenite is worse. Kenmare’s realised ilmenite price for the first half of 2026 was US$203 a tonne, down 29 per cent year on year, which Kenmare attributes to increased Chinese domestic supply and African imports. The one clearly strengthening price in the complex is zircon. Iluka has contracted a US$215 a tonne increase for the September 2026 quarter, taking it to US$1,760. Tronox reported zircon revenue up 43 per cent year on year and said supply remained constrained across the industry. Zircon is 3 per cent of Tusker’s assemblage. On demand, 2025 was weak and 2026 is an early and uneven recovery. Tronox reported titanium dioxide volumes up 18 per cent year on year in the June 2026 quarter, the highest since mid 2022, with pricing flat year on year and up 5 per cent sequentially. Iluka reported improving pigment operating rates but said feedstock purchasing activity remains cautious and that housing and construction demand continues to be mixed. Its kilns are still idle. That is the honest read. Pigment volumes are recovering, prices are only just moving, and feedstock demand is lagging both. 4.4 The forecast everybody quotes, and what it is worth The number in every rutile presentation is TZMI’s, that rutile demand from the titanium metals industry will grow about 3 per cent a year over the next decade while global supply declines about 7 per cent a year. We have been unable to verify that forecast at source. It reaches the public domain through investor facing media citing TZMI, and TZMI’s own studies are behind a paywall. We could not obtain the publication date of the underlying work. Wood Mackenzie and Project Blue rutile forecasts could not be sourced at all. So we say it plainly. There is no publicly verifiable, dated, quantified third party forecast of a rutile supply gap. The structural argument is real in direction, because Kwale is gone, Sierra Rutile is opaque and Richards Bay is sustaining rather than growing. The number that everybody repeats is not something a reader can check. And there is counter arithmetic that belongs in the same breath. Sovereign Metals’ Kasiya, at 222,000 tonnes a year of rutile, would be roughly half of current total world rutile production. One project could close the deficit on its own. A market this small can be re-supplied quickly, and that is as true a statement as the scarcity argument. 4.5 Cameroon The 2023 Mining Code, with implementing decrees enacted in November 2024, governs the sector. An exploration permit requires a mining agreement with the State. An industrial mining licence runs twenty years initially. The fiscal terms carry the headline risk. The State takes up to 10 per cent of share capital free and non dilutable, and may increase that to 35 per cent by mutual consent. The State’s production share for non precious minerals is set at 2 to 15 per cent, and the code does not fix the rate. It is negotiated. That is a very wide band and an unquantified cost for any explorer that gets as far as a mining agreement. Add a local capacity building levy of 0.5 to 1 per cent of turnover, a sector development fund and a site restoration fund, plus local content rules reserving 95 per cent of unskilled positions for nationals. On stability, we will not soften this. Paul Biya, aged 92 and in power since 1982, was declared winner of the October 2025 election with 53.66 per cent of the vote. The opposition candidate rejected the result and both men claimed victory. Protests followed across several cities. The death toll is disputed at between 16 and 48 depending on whose count you take, with more than 200 arrests. Ghost town protests in early November 2025 disrupted regional trade, with Douala port delays estimated at more than US$18 million a day. As at January 2026 the assessment from independent observers was that the crisis is far from resolved. Succession risk under a 93 year old president who began another seven year term in late 2025 is the dominant jurisdiction risk here and it is not one that can be modelled. Against that, Cameroon is trying. Mining is about 1 per cent of GDP today and the government has set a target of mining revenue overtaking oil, with several projects in commissioning including Minim Martap bauxite. In August 2026 it suspended an entire class of semi mechanised gold permits pending compliance and referred 137 illegal operators to court. A regulator willing to do that is either reassuring or alarming, depending entirely on which side of the licence you are standing. 4.6 Malawi The Mines and Minerals Act of 2023 sets a 5 per cent royalty on gross revenue and 30 per cent corporate tax. The resource rent tax is effectively suspended, with one mining development agreement describing it in its own text as not fit for purpose, and no replacement legislation had been enacted as at April 2026. Government equity is negotiated rather than statutory, with a 10 per cent non diluting interest as the precedent. Malawi’s statutory terms are materially lighter than Cameroon’s. They are also set by unpublished, individually negotiated agreements with no recorded parliamentary scrutiny, and a suspended windfall tax that could be legislated back at any time. Neither country is a low risk jurisdiction. They are risky in different ways, and a reader should not assume the Malawi remnant is the safe part of the portfolio. 4.7 A word about jurisdiction risk in Africa Before anyone uses the word Cameroon as a one word dismissal, it is worth saying something about jurisdiction risk on that continent generally, because the way it is usually handled in Australian small cap commentary is not much use to an investor. Africa has decades of mining progress behind it, and jurisdiction issues have been part of that entire history. They are not new, they are not confined to any one country, and they have not stopped mines being built. What they have in common is that they are an unknown factor. They can go left or they can go right. That is true of political images anywhere, and it is always not what it looks like and what it seems. The evidence in this very piece runs both ways, which is the point. Sierra Leone hosts the world’s largest natural rutile mine and has done for decades, through a civil war and everything that followed. Kenya’s Kwale mine ran from 2013 until December 2024 and then closed because the ore ran out, not because the country did anything to it. Mozambique has Kenmare at Moma, South Africa has Richards Bay, and Rio Tinto has just committed US$473 million to extend it to 2050. In Malawi, Rio Tinto holds close to a fifth of Sovereign Metals. These are not fringe operators taking wild risks. They are among the largest mining companies on earth choosing to be there. And it goes the other way too. In late July 2026 an International Chamber of Commerce tribunal awarded Sundance Resources Limited and its Cameroonian subsidiary Cam Iron SA approximately US$616 million in damages, interest and costs against the Republic of Cameroon, over Cameroon’s conduct in relation to the Mbalam iron ore project. Sundance’s own statement of the findings is that Cameroon breached its legal obligations to the two companies in relation to their investment, and breached the arbitration agreement by failing to comply with an emergency order issued against it in March 2022. Two things belong beside that, because the case is not the one sided warning it first appears to be. A separate tribunal dismissed Sundance’s parallel claim against the Republic of Congo in January 2026, in full, on a claim of US$8.8 billion. And the Cameroon award remains unpaid. Sundance has said that if Cameroon does not comply voluntarily, enforcement proceedings will be needed. Its litigation funder has said publicly that the matter remains subject to significant litigation and collection risk. Sundance itself was removed from the ASX official list in December 2020 after two years of suspension. That is the whole shape of jurisdiction risk in one case. A company can be right, can be found to be right, and can still be a long way from being paid. It is also a reminder that the process exists at all, and that a state can be held to it. So we are not going to tell a reader that Cameroon is fine, and we are not going to tell them it is hopeless. Neither statement would be honest and neither would be useful. What we will say is that jurisdiction risk in Africa is real, is permanent, is unknowable in direction, and has never by itself been a reason not to look. It is a reason to size a position accordingly and to expect the unexpected in both directions. Anybody who tells you they know which way a political situation on that continent will turn is telling you something they cannot know. 5 Where This Sits Against Its Peers We are on Track B, and we say so up front. No scoping study, no pre-feasibility study, no definitive feasibility study, and no JORC Mineral Resource exists on any Tusker asset. Under our own rules that means no valuation, no price and no per share range appear anywhere in this workup. What follows is relative positioning and a map of what would establish value. The comparison below is Samso’s own. We built it from published JORC resource and reserve statements, feasibility study announcements and market data, each cited with its date. Tusker has not endorsed it and did not supply it. One qualification governs every line. Tusker’s Douala Basin figure is a JORC Exploration Target, which is conceptual. Every other project in these tables carries a reported Mineral Resource or Ore Reserve and several are operating mines. Tusker will look cheap on any ratio for exactly that reason, and a reader who takes the ratio without the qualification will reach the wrong conclusion. 5.1 Grade, like for like Figure 7. In situ rutile grade, Douala Basin against its peers. The Douala Basin bar is a JORC Exploration Target, which is conceptual. Every other bar is a reported Mineral Resource or Ore Reserve. (Samso Research Media House.) Three readings, and we will give all three. Against Kasiya, Douala is about a third of the grade. Kasiya also carries 0.95 per cent graphite in its Mineral Resource as a second revenue stream. Douala’s second stream is kyanite, which has no assigned value. Against Sierra Rutile, Douala sits at or barely above the cut off. Sierra Rutile reported its mineral resources at cut off grades of 0.25 to 0.50 per cent rutile and its resource grades ran 0.55 to 1.60 per cent. A deposit averaging 0.30 to 0.35 per cent would, at Sierra Rutile’s own reporting cut offs, be marginal material rather than the payable core. That is the sharpest benchmark available and it comes from an audited resource statement. Against Kwale, Douala is about half the rutile grade. Kwale ran at 0.65 per cent rutile inside a 4.89 per cent heavy mineral ore where ilmenite and zircon carried substantial additional revenue. Douala’s heavy mineral grade is nominally lower and its payable fraction is lower again. 5.2 Douala against the large heavy mineral sands projects Comparisons of Douala Basin against the world’s large rutile and heavy mineral sands projects are circulating. We have rebuilt that comparison from each company’s current resource and reserve statements rather than repeating the figures in circulation, because several of those figures are superseded and one is not supported by a JORC table at all. What the table actually shows, and what it cannot The classification column is the one to read first. Douala is the only exploration target in the set. Every other line is a reported Mineral Resource or Ore Reserve, and four of them belong to projects that are either operating or through feasibility. That is not a like for like comparison and no amount of arranging the rows makes it one. The category mismatch inside the table matters as much as the one at the top. Kasiya carries 20.24 Mt of contained rutile in its Mineral Resource and 5.09 Mt in its Ore Reserve. That is a fourfold difference for the same deposit, and it is the difference between what is in the ground and what the company has committed to mining. Moma reports its resource excluding reserves, so the two lines add rather than overlap. Copi has no ore reserve at all. Any single number comparison across these projects is choosing between incompatible categories. The most useful thing in the table is the last column, and it is not the tonnage. Of the six projects, only three report a rutile grade as rutile. Sovereign does, and rutile is the product. Kenmare does, at 0.05 per cent, and rutile is under one per cent of what Moma actually ships. Tusker’s rutile figure is derived from a modelled assemblage rather than measured. Thunderbird reports no rutile at all, only HiTi leucoxene. Copi reports rutile combined with HiTi85 and cannot be split. Donald reports rutile combined with anatase. So the honest reading is this. On tonnage, Douala sits in the same range as Kasiya, Copi and Donald, and that is real. On grade, it is roughly a third of Kasiya and it is the only figure in the column that is conceptual. And on the assemblage, it is one of the few in the set where a rutile number exists at all, which cuts in the company’s favour, and it has not been drilled out, which does not. 5.3 The comparison that is harsher, and fairer A mineral sands operation is paid for what it recovers, not for what it digs. So the more useful measure is the in situ grade of minerals someone will actually buy. Samso Research Media House calculation, using each company’s published assemblage percentages applied to its published heavy mineral grade. Douala’s saleable fraction is taken as rutile at 15 per cent plus zircon at 3 per cent of total heavy minerals, excluding kyanite and ilmenite. Working shown so a reader can check it. Figure 8. In situ valuable heavy minerals, being the fraction of the rock that has an established market. This is the comparison that decides whether a heavy mineral grade is worth anything. (Samso Research Media House.) On this measure Douala sits at about half the grade of Kwale's residual reserve, which is the depleted tail of a mine that has already closed, and at about an eighth of Thunderbird and Ranobe. 5.4 A revenue test, with the assumptions named This calculation is ours and its assumptions are ours. We show it because it is the most direct way to explain why grade matters so much in this business. Kasiya at full production of 24 million tonnes a year would generate about US$725 million in annual revenue from rutile and graphite together, which is about US$30.20 per tonne of ore mined. Rutile alone is about US$15.45 per tonne of ore. Its operating cost, at US$450 per tonne of product, works out at about US$9.32 per tonne of ore. Douala at the midpoint grade, rutile only, at the same rutile price, and assuming 80 per cent recovery, which is our assumption and is not sourced from any testwork, gives 2.56 kilograms of rutile per tonne of ore. At US$1,670 a tonne, which is the rutile price the Kasiya definitive feasibility study assumes, that is about US$4.28 per tonne of ore. At Iluka's realised US$1,254 a tonne from Section 4.3 it is about US$3.21. So gross rutile revenue of US$3.21 to US$4.28 against a benchmark operating cost of US$9.32, before any capital at all. Zircon would add something. Kyanite has no verifiable price at this scale. Recovery may be better or worse than 80 per cent, because no testwork has been done. We are not saying the project cannot work, and this calculation is not a verdict on it. It is a way of showing why grade decides everything in this business. Run at the modelled target grade with our assumptions, the rutile stream on its own does not clear a benchmark operating cost. Run it at the reconnaissance grades and the picture is different again. Which of those two the deposit actually is has not been established, and that is the open question, not a conclusion. 5.5 Where the market has it priced The most instructive line in that table is Lion Rock. It holds Cameroon rutile ground adjacent to Tusker’s. It has no JORC resource and no exploration target at all. Its reconnaissance drilling averaged about 0.8 per cent in situ rutile across 107 test locations, which is roughly two and a half times Tusker’s modelled target grade. Tronox holds about 5 per cent of it. The market values it at about 8.5 times Tusker’s market capitalisation. A reader can take that two ways. Either Tusker is cheap against the nearest thing to a direct peer, or the market has looked at both sets of grades and priced them accordingly. We think the grade difference explains most of the gap, and that the absence of any resource at either company explains why neither is worth more. It is also worth noting that Tusker’s market capitalisation sits among generic African nano cap explorers rather than among rutile stories. MRG Metals, at a comparable market capitalisation, holds a reported 860 million tonne mineral resource in Mozambique. That is what the market currently thinks a conceptual target is worth. 5.6 On the possibility of a re-rating, and what usually happens We do not publish price targets and we are not going to start here. But readers of small caps think in multiples of market capitalisation, so we will deal with it directly and with the evidence on both sides. The precedents exist and two of them are in this exact commodity and country. Peak Minerals, now Lion Rock, rose about 430 per cent during 2025 to a market capitalisation of roughly A$134 million on the Minta rutile discovery, with no resource and no exploration target at the time. Sovereign Metals went from about US$25 million in 2019 to about US$110 million in 2020 and about US$170 million in 2021 as Kasiya was defined, and it held that re-rating. And DY6, which is this company, rose about 475 per cent through 2025 to a market capitalisation of about A$17 million on the Cameroon acquisition. That last one is the point. Tusker’s re-rating already happened, and the share price has given back roughly 78 per cent of it from the high. The market capitalisation has fallen less, about half, because shares were issued on the way down. Lion Rock’s share price is down about 70 per cent over twelve months on the same basis. Anyone buying this today for a five bagger is counting on a second re-rating, because the first one has already happened and been given back. The counterweight has to be stated with equal force, because most explorers do not do this. Strandline Resources went into administration in February 2025, into liquidation in August 2025, and was delisted, after building a mine on a A$150 million government backed facility. Walkabout Resources followed a similar path. Image Resources is down about 76 per cent and Sheffield about 72 per cent, and both are producers. MRG Metals declared a large high grade inferred resource in February 2022 and four and a half years later carries a A$10.5 million market capitalisation. And Sierra Rutile, the world’s largest natural rutile producer, was taken private at about A$76 million. Being the global number one in this commodity did not produce a re-rating. It produced an exit. Across the names we surveyed for this workup, the scorecard is one sustained re-rating, two five baggers that gave back 70 to 78 per cent within about a year, two total losses, two producers down more than 70 per cent, and one large resource dormant for four years. This is context for the call. It is not the call. 6 What Would Change Our Mind What would make us more positive Infill drilling that repeats the reconnaissance grades over area. This matters more than anything else on the list. The seven reconnaissance holes ran 0.33 to 0.76 per cent rutile across eight intervals, well above the 0.30 to 0.35 per cent modelled target, and they sit outside the target footprint. If systematic drilling turns that into a continuous higher grade zone, the whole assessment above changes, because the argument in Section 5 is against the modelled grade, not against the drilled one. A maiden JORC Mineral Resource at Douala Basin. A resource with a meaningful Indicated component would move this company out of the conceptual category entirely and make every comparison in Section 5 a fair fight rather than a mismatch. Note the company has guided six to nine months for the exploration work, not for the resource, and has said no assurance can be given that a resource will be defined. Metallurgical testwork showing good recovery and a saleable rutile specification. There is none. Any result at all is new information. A kyanite product specification with a named buyer. This would reframe 70 per cent of the assemblage from waste to revenue and it is the highest leverage single item on the list. Completion of the Green Exploration sale, on time. It funds the drilling without dilution at a record low share price. Grant of the Central Rutile and Yaounde West permits. Turning applications into ground removes a risk most readers do not know is there. The questions we cannot answer yet What follows is not a bear case. We do not have enough to make one, and we are not going to build a verdict out of gaps in the record. These are the questions we would want answered before anyone treats this as more than a watch, set out with the facts we do have and an honest statement of where those facts stop. Is the modelled grade the real grade? The target is modelled at 0.30 to 0.35 per cent rutile. The seven reconnaissance holes returned 0.33 to 0.76 per cent across eight intervals, and they sit outside the target area. Those two numbers describe different things and we cannot reconcile them from the outside. Systematic drilling will. Until it does, nobody knows which one describes the deposit, and that includes the company. Does the kyanite have a market? Seventy per cent of the heavy mineral assemblage is kyanite, on the company’s own disclosure, and the company has said its recovery, product quality and commercial potential all remain subject to further testwork and market assessment. No testwork has been published. We have set out in Section 3 what the global kyanite market currently looks like and what the arithmetic implies if the kyanite is not saleable. We cannot tell a reader whether it is, because nobody has tested it. Will the ground convert, and on what terms? Twenty one Central Rutile permits are applications rather than grants, and the Yaounde West licences had not been announced as granted at the time of writing. Cameroon’s code sets the State’s production share for non precious minerals at 2 to 15 per cent and leaves the rate to negotiation. We do not know when the permits convert or what they will cost when they do. What happens in Cameroon next? A 93 year old president began a further seven year term in late 2025 after a disputed election, with deaths and arrests in the aftermath and no succession settled. We will not forecast that, and we would be sceptical of anyone who does. Section 4.7 sets out how we think about jurisdiction risk in Africa generally, which is that it is real, permanent, unknowable in direction, and has never by itself been a reason not to look. It sits underneath every asset in the Cameroon portfolio and each reader should size it to their own tolerance. Does the funding arrive in time? Cash was A$1.35 million at 30 June 2026 against roughly A$5.1 million of all in spending over the prior twelve months. The Green Exploration sale to AuKing is binding but had not completed, with a long stop date of 30 September 2026. Whether it completes on time, and what the company does if it does not, are both unknown to us today. What does the metallurgy say? There is no published metallurgical testwork on any Tusker asset. Not at Douala Basin, not at Central Rutile, not at Mzimba. Testwork commissioned at Tundulu in April 2025 was never reported on publicly. Recovery and product specification are what turn a grade into revenue, and we have nothing to work with on either. Has the strategy settled? The company has changed what it is twice in three years and has sold or agreed to sell almost everything it listed with in 2023. Whether that reads as a company that found its focus or a company still looking for it is a fair question, and we do not think three years of history is enough to answer it either way. Every one of those is answerable. None of them is answered today. That is the reason for the call in Section 8, and it is the whole reason for it. The settling event Of all of those questions, one event answers the one that matters most. The maiden JORC Mineral Resource at the Douala Basin project. There is no date to put on it. The company has guided approximately six to nine months from July 2026 for the exploration activities that test the target, being drilling, density work, metallurgy and modelling, subject to permitting and assay turnaround. It has not guided a date for a resource, and it has said in terms that no assurance can be given that one will be defined. So the earliest a resource could realistically follow is some time after the first half of 2027, and it may not come at all. Assay results in the last two quarters have also run later than flagged. The Green Exploration completion on 30 September 2026 is nearer and more binary, and it matters, but it settles whether Tusker can afford to keep going rather than whether the deposit is worth going after. That is why it sits in the watch list rather than under the call. 7 Management and the DNA Question The question Samso always asks is whether the people running a company have taken something the whole way before. On this board the answer is narrow but not empty. Plate 1. The Tusker Minerals board and senior technical team as at the time of writing. Daniel Smith, Myles Campion, John Kay, Dr Nannan He, Cliff Fitzhenry and Troth Saindi. Source. Portraits from the company website, tuskerminerals.com, About Tusker Minerals page, accessed 2 September 2026. Panel assembled by Samso. Nobody on this board has taken a project to financing or production. Two people have taken one a long way, and neither did it as the accountable decision maker at Tusker’s level. Cliff Fitzhenry, chief executive since July 2025. MSc and BSc Honours from the University of KwaZulu-Natal, registered professional natural scientist, based in South Africa. About twenty years of exploration, principally in Africa. He was senior geologist at Sovereign Metals leading in country exploration on Kasiya, which became the world’s largest natural rutile deposit. That is the strongest technical credential on the team and it is exactly on point, because the deposit style and the analogue are the same ones Tusker is chasing. The qualification is that he was the geologist, not the person who financed or developed it, and there is no evidence he has run a company through a resource, a study or a financing. Daniel Smith, executive chairman since November 2022. A capital markets and governance professional who founded the Perth corporate services firm Minerva Corporate in 2013 and has been involved in more than a dozen listings across the ASX, AIM and NSX. He holds board or company secretarial roles at a number of other small caps. Across that record we found no instance of a project taken to resource, study, financing or production under his stewardship. Company formation, listing and board service across many small explorers at once is a real skill. It is not development experience, and readers should understand the difference. Myles Campion, non executive director since March 2023. MSc from the Royal School of Mines, thirty years across exploration geology and fund management, including project geologist work at LionOre on the Emily Ann nickel mine. This is the one genuine credential of the type we are looking for. As executive chairman and chief executive of Europa Metals he took the Toral zinc, lead and silver project in Spain through a JORC resource, a resource upgrade and a scoping study. We report the outcome as well, because that is the point of asking the question. Europa did not take Toral further. It sold the project to Denarius Metals in 2023, spent six months looking for a replacement asset, found candidates but did not secure funding for a reverse takeover, was suspended from AIM in May 2025 and resolved to return its assets to shareholders. That is a failure to finance rather than a governance failure, and it is directly relevant to a company whose plan requires funding a drill out. John Kay, non executive director and company secretary. A corporate and commercial lawyer, co-founder of the advisory firm Arcadia Corporate, with seventeen years or more in equity capital markets and resources. Corporate function, no technical background, no development record. Dr Nannan He, non executive director. PhD in geochemistry from Curtin, degrees from Hokkaido and Jilin universities, and manages an investment vehicle looking at exploration and resource projects. Technically qualified, but the role is investment rather than delivery. We note that Zhenshi Group, a Hong Kong composite materials group, was a A$1.5 million cornerstone investor at the 2023 IPO and remains a substantial holder. We could not source any employment or nominee relationship between Dr He and Zhenshi, and we are not asserting one. Troth Saindi, exploration manager for Africa. MSc from the University of the Witwatersrand, registered competent person, and seventeen years spanning exploration through to production, including a decade as group geologist at Bushveld Minerals while it was an operating vanadium producer. Genuine production side exposure is rare on a team this size and it is a real positive. The piece of history that answers the question best In August 2023 the company hired Lloyd Kaiser as chief executive. He had ten years at Arafura Rare Earths, as general manager of sales and marketing negotiating rare earth offtakes and as general manager of technology and engineering on the Nolans Bore project. He was the one executive with downstream development and offtake experience. He was gone by July 2025, replaced by an exploration geologist, and the project he was hired to develop was agreed for sale in August 2026. That is not a criticism of Cliff Fitzhenry, who is well suited to what this company now does. It is an observation about what this company decided it was. It went from trying to develop something to trying to find something. Those need different people and Tusker changed its people accordingly. Whether that was the right decision is exactly the argument this workup leaves open. 8 Samso Concluding Comments SAMSO CALL: WATCH AND DYOR The settling event we are watching is the maiden JORC Mineral Resource at the Douala Basin project. There is a version of this story that is very easy to tell. A tiny company with an A$8 million market capitalisation is sitting on a multi billion tonne titanium target, within about 40 kilometres of a deep water port, in a country that produced high purity rutile eighty years ago and has barely been looked at since. The world’s largest rutile mine has been taken private and no longer reports. A producing mine in Kenya has closed. Nobody is building new supply. Tell it that way and you do not need a spreadsheet. We do not think that version is dishonest. We think it is incomplete, and the missing part is not a counter-argument. It is a list of things nobody knows yet. We do not know whether the modelled grade of 0.30 to 0.35 per cent or the reconnaissance grades of 0.33 to 0.76 per cent describe this deposit. We do not know whether the kyanite, which is seventy per cent of the assemblage, is product or waste, because it has not been tested. We do not know when the flagship permits convert from applications to grants, or on what fiscal terms. We do not know what the metallurgy says, anywhere in the portfolio, because none has been published. We do not know whether the Malawi sale completes by 30 September. And we will not pretend to know what happens next politically in Cameroon. Set against that, the thing that keeps us interested is small but it is real. The seven reconnaissance holes came in at up to two and a half times the modelled grade, and they were drilled outside the target area. Either the target understates the deposit or those holes are the good bits. Nobody knows yet, including the company, and that is precisely why it is worth watching rather than acting on. So the call is Watch and DYOR. Not because the story is bad, and not because we are hedging. Because the single number this entire proposition rests on, the grade over area, has not been established. The work that tests it is guided at six to nine months. A resource, if one comes at all, sits somewhere beyond that, and the company has been careful not to promise it. What we would say to a reader is this. If you already own it, the near term question is not the grade, because the grade will not be settled for months. It is whether the company can fund itself as far as the resource. Watch 30 September. If you do not own it and the scale is what draws you, understand that what you are buying today is a conceptual number rather than an established one, and that the nearest comparable in the same country, on better reconnaissance grades and with no resource either, has still lost about 70 per cent of its value this year. Make of that what you will. We are not going to tell you what it means. Look at the facts. Take a balanced view. And then decide for yourself whether the drilling that comes next is worth waiting for. 9 Disclaimer The information contained in this Samso Research workup is general in nature and does not take into account the objectives, financial situation or needs of any particular person. It is not financial product advice. Figures are drawn from public sources believed reliable at the stated dates and are not guaranteed. Market sensitive numbers change daily. Readers should seek professional advice before making any investment decision. The Samso Way – Seek the Research Here at Samso, we pride ourselves on delivering content for investors that is independent and informed by over three decades of experience in the industry. Our content is well-researched and is only created if I see merit in discussing the company's story. Our mission is simple. Cut through the noise and spotlight what matters. Genuine stories, grounded insights, and real opportunity. Investors can explore our four core platforms: Coffee with Samso Samso Insights Samso News Samso Research There may be numerous paths to success in investing, but the common thread among successful individuals is that they remain committed to making informed decisions. Equip yourself with the right knowledge and tools, and you will be well on your way to achieving your financial goals. The Samso Philosophy: Stay curious. Stay sharp. And remember, digging deeper always uncovers the real value. Happy Investing, and the only four-letter word you need to know is DYOR. Disclaimer This Samso Research Workup is independent research commentary for general information only. Nothing in this document is financial product advice, and it does not take into account any reader’s objectives, financial situation or needs. Figures are drawn from public sources believed reliable at the stated dates but are not guaranteed. Market-sensitive numbers change daily. Samso or associated parties may hold positions in, or have commercial arrangements with, companies mentioned. Seek professional advice before making investment decisions. Read full Disclaimer. About Samso Samso is a trusted platform that equips dedicated investors with up-to-date industry knowledge and insights from top CEOs and thought leaders. Samso Research | www.samso.com.au | An Investor Lens on ASX-Listed Companies

Part 1 argued that the world is spending enormous sums to produce AI, whether or not that spending yet buys value. This part follows the money into the physical supply chain the spending pays for, the chips and the buildings, and asks a plain question: which of it can an Australian investor actually own on the ASX, and which of it, including the much-hyped vector databases, is simply not listed here at all. WHAT THIS PIECE COVERS 1.00 THE PICKS-AND-SHOVELS IDEA, AND ITS LIMIT Selling shovels in a token rush, and where the analogy breaks. 2.00 THE CHIP LAYER: SEMICONDUCTORS ON THE ASX What is actually listed here, and how much of it is selling chips. 3.00 DATA STORAGE FOR AI: THE DATA CENTRE, NOT THE DATABASE Where AI really keeps its data, and which of that is investable here. 4.00 HOW THE TWO HALVES CONNECT Why the cost-per-outcome test applies to the shovels as well. 1.00 — THE PICKS-AND-SHOVELS IDEA, AND ITS LIMIT Selling shovels in a token rush There is an old rule about gold rushes: the steadier money was often made selling picks, shovels and denim to the miners rather than digging for the metal. The equivalent question for artificial intelligence is which businesses get paid no matter which AI application eventually wins. Part 1 made the case that a company's AI spending should be judged on cost per outcome, the cost of a finished, useful result, rather than on how much it spends or how loudly it talks about AI. That test is deliberately hard on the buyers of AI. The suppliers are a different proposition. Whether or not a given bank or retailer ever proves its chatbot was worth it, the tokens (the small chunks of text an AI reads and writes, and the unit it is billed by) still had to be produced somewhere, on physical chips, inside physical buildings, drawing physical power. FIG. 01 lays out that supply chain as a stack, and marks the plain truth of it: an ASX investor can buy the bottom of the stack (the buildings and, speculatively, the chips) but not the fashionable middle (the models and the vector databases). Now the limit, stated before any company is named. Australia is a small part of this supply chain. It has no large-scale chip fabrication, so most of its listed semiconductor names are early-stage technology-licensing hopefuls rather than profitable manufacturers. What follows maps what is here, what it does, and what each name's main risk is. It is not a list of recommendations, and the counterweights are as important as the descriptions. 2.00 — THE CHIP LAYER: SEMICONDUCTORS ON THE ASX Small, speculative, and mostly selling ideas rather than chips A semiconductor is the material, usually silicon, from which computer chips are made, and "the semiconductor industry" is the chain that turns it into working processors and memory. That chain has several rungs, and the handful of ASX names sit on different ones. FIG. 02 places them. The names differ as much in maturity as in what they do. FIG. 03 maps them on two axes an investor cares about: how far each is from real, repeatable revenue, and how speculative it remains. Most cluster at the pre-revenue, high-risk corner. Only DUG Technology and, to a lesser extent, Revasum sit near the "actual revenue" end. Before the profiles, PLATE 3 is a reminder of what this whole layer eventually produces: physical chips. Design and intellectual property BRN · AXE BrainChip Holdings (ASX: BRN) designs a low-power processor called Akida for "edge" AI, meaning AI that runs on a device such as a drone, sensor or piece of defence equipment rather than in a distant data centre. It licenses the design to others. For the year to December 2025, revenue was about US$1.89 million (up from about US$0.4 million a year earlier) against a net loss of about US$20 million, and a next-generation processor (the AKD2500) is expected in prototype during 2026. Risk Revenue is minimal, the licensing model is unproven at commercial scale, and the company has burned cash for years. The share price reflects hope for the future, not current earnings. PLATE 1 shows the Akida design BrainChip licenses to others, with its on-chip processing blocks and "neuron fabric". Archer Materials (ASX: AXE) is a research-stage developer of a quantum computing chip and a separate biochip. Quantum computing is an experimental approach that could one day solve certain problems far faster than today's machines. Archer reported a net loss of about A$2.99 million for the half and generates almost no revenue. Risk Highly speculative, pre-revenue, and any commercialisation is years away and uncertain. Memory WBT · 4DS Both names are chasing the same prize shown in PLATE 2: a memory that is as fast as today's working memory but keeps its data without power. Weebit Nano (ASX: WBT) has developed a memory technology called ReRAM (resistive memory) that aims to be faster and more power-efficient than the Flash memory in today's devices, and licenses it to chipmakers. It reported a record half-year revenue of about A$5.6 million, referenced a licensing deal with Texas Instruments, held about A$82.8 million in cash, and guided to more than A$10 million of revenue in the 2026 financial year. Its market value is above A$1 billion. Risk The valuation rests on future royalties that are still ramping, not on current profits, so it is priced for success that has not yet arrived. 4DS Memory (ASX: 4DS) is developing its own version of resistive memory aimed at high-density storage for mobile devices and data centres, which makes it the closest chip-level name to the "data storage for AI" theme. Revenue is negligible at about A$316,000, with roughly A$3.34 million in cash plus about A$5.45 million in deposits. Risk Very early stage with almost no revenue; it needs a commercial manufacturing partner before it counts for much. Materials BLG · SLX BluGlass (ASX: BLG) makes laser components from gallium nitride, a semiconductor material used in lasers and power electronics, using its own manufacturing process. It reported revenue of about A$5.76 million (up about 14 per cent), a net loss of about A$4.10 million, and referenced a US defence contract and a project pipeline. Risk Loss-making and facing the challenge of scaling a specialised process to volume. SILEX Systems (ASX: SLX) belongs here only with a clear caveat. Its main business is laser uranium enrichment technology, not chips. Its semiconductor-relevant activity is a "Quantum Silicon" plant producing highly purified silicon-28, a material used to build quantum computing chips. That is a genuine but secondary part of the company. Risk This is primarily an enrichment-technology company; anyone buying it for semiconductor exposure is really buying the enrichment story with a small chip option attached. Equipment and high-performance computing RVS · DUG Revasum (ASX: RVS) is the most literal "shovels" name on this list. It makes the grinding, polishing and planarising machines used to manufacture semiconductor wafers, including silicon carbide, and so sells tools to the industry rather than betting on any single chip winning. Risk A small supplier into a cyclical capital-equipment market; demand rises and falls with chipmakers' own spending. Current financials should be checked directly. DUG Technology (ASX: DUG) is different again: an established provider of high-performance computing (very large-scale number crunching), cloud services, geoscience work and software. Unlike the chip hopefuls it earns real revenue, and reported strong growth with its high-performance computing demand rising and cash flow turning positive. Risk Small in a field where the giants of cloud computing set the price, so it must keep finding niches they do not serve. 3.00 — DATA STORAGE FOR AI: THE DATA CENTRE, NOT THE DATABASE The vector database everyone talks about is not listed here Ask where AI keeps its data and the fashionable answer is the vector database. The realistic answer for an ASX investor is that you cannot buy one here, so the real local exposure is the building the data sits in. First, what a vector database is, in plain terms. Traditional databases find things by exact match, the way a phone book finds a name you spell correctly. A vector database stores information as strings of numbers that capture meaning, so it can find things that are similar in sense rather than identical in spelling. It is the technology behind most "chat with your documents" features, where an AI answers using a company's own material. It is central to how businesses are putting AI to work. It is also not something you can own on the ASX. The specialist vector database companies are privately held, and the listed ways to touch the technology are large overseas firms that offer it as one feature among many. One of the more prominent independents, Marqo, has Australian roots but is private (a point worth confirming before relying on it). FIG. 04 sets out where the category actually trades. So the ASX way to own "storage for AI" is the physical layer: the data centre. A data centre is a large, secure, heavily powered building full of computers that store data and run software for other companies. Its size is measured in megawatts (MW) of power, because power, not floor space, is what limits how much AI computing a site can run. FIG. 05 explains how a data centre turns power into profit, and where the cost-per-outcome test from Part 1 bites. Australia has a genuine, investable cluster of listed operators. FIG. 06 compares the capacity figures the main names have disclosed, with an important caveat in the caption about how those figures are defined. The listed data-centre names NXT · MAQ · DGT · GMG · IFT NextDC (ASX: NXT) is described as the purest data centre business on the ASX: it builds and runs the buildings that house AI computing and storage. It referenced 667 megawatts of contracted capacity (as at April 2026, up from 416.6 megawatts in December 2025), a A$1.5 billion equity entitlement offer forming part of a A$2.2 billion capital plan announced in April 2026, and a partnership with OpenAI to build sovereign AI infrastructure in Sydney (data centres kept on Australian soil under Australian control). Risk It spends enormous sums upfront; returns depend on filling that capacity with paying tenants at prices above its cost of capital. PLATE 4 is an artist's impression of NextDC's S4 Sydney campus, one of the hyperscale sites behind its contracted capacity. Macquarie Technology (ASX: MAQ) runs data centres alongside cloud and telecommunications businesses, and hosts a large share of Australian Federal Government computing, which is a demanding, sticky customer. A new 47 megawatt facility was referenced for opening in September 2026, with more in the pipeline. Its market value is around A$1.7 billion. Risk Smaller than NextDC, so expansion carries more execution and funding risk relative to its size. DigiCo Infrastructure REIT (ASX: DGT) is a data centre real estate investment trust (a listed property trust). In May 2026 it announced the sale of its Chicago data centre for about US$750 million to strengthen its balance sheet and fund growth, including an approved 88 megawatt expansion of its SYD1 Sydney campus. It has traded below its A$5.00 December 2024 listing price. Risk Recently listed and already below its issue price; as a property trust it is sensitive to interest rates. Goodman Group (ASX: GMG) is a large industrial property group that has pivoted hard towards data centres, referencing a development pipeline of about A$14.4 billion that is roughly 73 per cent weighted to them, and multi-gigawatt power capacity. It is the most diversified, and therefore lower-risk, way to touch the theme. Risk Data centres are only a slice of a business still exposed to the broader property cycle, so the AI exposure is diluted. PLATE 5 shows one of Goodman's data centre developments, the kind of asset its pipeline is now weighted towards. Infratil (ASX: IFT) owns about half of CDC Data Centres, a major operator that referenced a 555 megawatt contract with a large cloud customer, taking its contracted capacity above one gigawatt. Infratil is a New Zealand-based, dual-listed infrastructure investor. Risk A diversified infrastructure owner, so the AI exposure is real but indirect and shared with airports, energy and other assets. The connective tissue MP1 Megaport (ASX: MP1) is not storage, but it is worth naming as the plumbing between clouds and data centres. It sells "network-as-a-service", on-demand connections that let businesses link up cloud and data-centre services, and has been positioning towards AI workloads. Risk It provides connectivity rather than storage or compute, and competes in a fast-moving market; include it as an adjacency, not a core storage play. 4.00 — HOW THE TWO HALVES CONNECT The cost-per-outcome test applies to the shovels too It would be easy to read Part 1 and Part 2 as opposites: one urging caution on AI spending, the other pointing at ways to buy into it. They are the same discipline applied twice. The test from Part 1 was cost per outcome, not inputs. Turn it on the picks-and-shovels names and it still bites. A data centre operator is worth it when it signs paying tenants at a return above its cost of capital, not when it announces a headline number of gigawatts. A chip hopeful is worth it when it converts its intellectual property into recurring royalties, not when it runs an impressive demonstration. A high-performance computing provider is worth it when it wins work the cloud giants cannot serve more cheaply. In each case the announcement is the bill; the signed contract or the royalty cheque is the value. There is also a real tension between the two parts that a clear-eyed investor should hold. Part 1's cost discipline, if it takes hold, would slow the growth in tokens per task, which is part of what drives demand for all these shovels. The counter is that adoption has been growing faster than efficiency: Google's token volumes rose roughly seven-fold in a year even as the cost per token collapsed. That pattern, more use swamping cheaper units, is the bull case for the infrastructure. It is a reasonable bet, not a law, and Part 1's bearbox applies here too. References & sources Company activities and figures below are drawn from the secondary sources listed and are as at the companies' most recently reported results referenced there, broadly the first half of the 2026 financial year. Every company figure is market-sensitive and should be confirmed against each company's own ASX announcements and refreshed on publication day. All figures shown in the visuals are original Samso illustrations of the data named in each caption. Figures in this note were checked against the companies’ own ASX presentations and announcements where available (to July 2026); the NextDC OpenAI partnership, DigiCo’s US$750 million Chicago sale, Weebit’s Texas Instruments licence, Goodman’s pipeline, Macquarie’s 47MW facility and Infratil’s 555MW CDC contract were all confirmed against primary sources. This note names companies to illustrate where exposure exists; it is not a recommendation on any of them. Photographs and company diagrams labelled PLATE are reproduced from the companies' public ASX release presentations, with attribution in each caption, and are distinct from the original Samso illustrations labelled FIG. Veye — "Top 5 ASX semiconductor stocks 2026": Weebit Nano (WBT), Archer Materials (AXE), BluGlass (BLG), 4DS Memory (4DS), BrainChip (BRN) descriptions and half-year figures. Stocks Down Under — "Marvell... 3 ASX AI chip stocks WBT AXE BRN" and related coverage: Weebit Nano, Archer Materials and BrainChip technology and status; BrainChip market value and AKD-series roadmap. Stockhead — "Revasum leads a pack of ASX semiconductor stocks...": Revasum (RVS) grinding, polishing and CMP wafer equipment; SILEX (SLX), BrainChip, Weebit Nano, 4DS and BluGlass activity summaries. Silex Systems (SLX) company materials and Stocks Down Under coverage: primary business in laser uranium enrichment technology; "Quantum Silicon" (Q-Si) plant producing enriched silicon-28 for quantum computing. DUG Technology (DUG) FY26 half-year results release and Stocks Down Under coverage: high-performance computing, cloud and geoscience services; revenue growth and cash flow. Stocks Down Under — "The ASX AI infrastructure boom... data centre stocks": NextDC (NXT) 667 MW contracted, A$1.5 bn raise and OpenAI partnership; Macquarie Technology (MAQ) government hosting and 47 MW facility; DigiCo (DGT) Chicago sale and Sydney project; Goodman Group (GMG) pipeline and power capacity; Infratil (IFT) / CDC 555 MW contract and 1 GW-plus contracted capacity. Kalkine and Megaport (MP1) investor materials: network-as-a-service cloud connectivity and AI-workload positioning. AI Magazine — "Top 10 vector databases for AI"; MarketsandMarkets vector database market overview: specialist vector database vendors (Pinecone, Weaviate, Qdrant, Milvus/Zilliz, Marqo) and listed providers offering vector search as a feature (MongoDB, Microsoft, Amazon); no ASX-listed pure-play identified. Reproduced company visuals (PLATES): BrainChip Holdings 2026 AGM presentation (6 May 2026); 4DS Memory Capital Raising Presentation (January 2025); NextDC Investor Presentation (20 April 2026); Goodman Group 1H FY26 Results Presentation (19 February 2026). Each reproduced with attribution in its caption. Cover photo: Silicon wafer researcher, U.S. Department of Energy. Public domain, via Wikimedia Commons (https://commons.wikimedia.org/wiki/File:Silicon_wafer_researcher.jpg). Reproduced photograph, shown faded under a navy overlay on the article thumbnail. The Samso Way – Seek the Research Here at Samso, we pride ourselves on delivering content for investors that is independent and informed by over three decades of experience in the industry. Our content is well-researched and is only created if I see merit in discussing the company's story. Our mission is simple: cut through the noise and spotlight what matters—genuine stories, grounded insights, and real opportunity. Our content is well-researched and is only created if the team sees merit in discussing the company or concept. Investors can explore our three core platforms: Coffee with Samso Samso Insights Samso News There may be numerous paths to success in investing, but the common thread among successful individuals is that they remain committed to making informed decisions. Equip yourself with the right knowledge and tools, and you will be well on your way to achieving your financial goals. Most importantly, investors need to be absolutely diligent in understanding their own risk-reward tolerance and capabilities. Never bite off more than you can chew. As they say, Rome wasn't built in a day, and the Great Wall stood because it took centuries to complete. The Samso Philosophy: Stay curious. Stay sharp. And remember—digging deeper always uncovers the real value. In Life, there is no such thing as a Free Lunch. Never bite off more than you can chew is my parting comment. Happy Investing, and the only four-letter word you need to know is DYOR. To support our independent work, please head over to our Support Page and give us a helping hand in any of the ways listed. This is a new initiative for the Samso Platform, and it was always the concept of Samso when we started this journey in 2018. Disclaimer The information or opinions provided herein do not constitute investment advice, an offer, or solicitation to subscribe for, purchase, or sell the investment product(s) mentioned herein. It does not take into consideration, nor have any regard to your specific investment objectives, financial situation, risk profile, tax position and particular, or unique needs and constraints. Read full Disclaimer. About Samso Samso is a trusted platform that equips dedicated investors with up-to-date industry knowledge and insights from top CEOs and thought leaders. By staying informed on business advancements and market trends, investors can enhance their financial decisions through a combination of expert guidance and their own research. Samso Insights | www.samso.com.au | An Investor Lens on ASX-Listed Companies

I use AI in my own work, and I can see how far it extends what one person can do. But when this industry talks about itself, the words get in the way. This Insight takes one real presentation by a leading AI company, quotes the jargon exactly as spoken, and translates it piece by piece, so that by the end you can read any AI announcement and know what is actually being said. WHAT THIS PIECE COVERS 1.00 WHY I AM WRITING THIS The language got between me and the industry. 2.00 THE UMBRELLA WORDS: AI, MACHINE LEARNING AND WHAT AN LLM ACTUALLY IS Five nested words, and the engine sitting behind the app. 3.00 THE TOKEN: THE UNIT EVERYTHING ELSE IS PRICED IN A small piece of a word, and the industry's whole accounting system. 4.00 TRAINING AND INFERENCE: BUILDING THE PLANT AND RUNNING IT The two lives of a model, and the words that belong to each. 5.00 SCALING LAWS, TOKEN EFFICIENCY AND THE DATA WALL Why bigger has rules, and why the data is running out. 6.00 CONTEXT WINDOWS: HOW MUCH THE MODEL CAN HOLD IN ITS HEAD Working memory, measured in tokens. 7.00 GETTING ANSWERS OUT: PROMPTS, REASONING, MULTIMODAL AND LOOKING THINGS UP The words you meet when you actually use the tools. 8.00 AGENTS AND AGENT SWARMS: FROM ANSWERING TO DOING The industry's favourite word of 2026, explained plainly. 9.00 THE WORDS OF THE RACE: OPEN WEIGHTS, FRONTIER, AGI AND THE REST The vocabulary of ambition, and what it is worth. 10.00 FIVE QUESTIONS TO ASK OF ANY AI ANNOUNCEMENT The vocabulary, put to work on a real announcement. 1.00 — WHY I AM WRITING THIS The language got between me and the industry - The AI Token Mystery I am writing this Insight for a simple reason: the language of AI still confuses me. I use AI, and I can see how far it extends what one person can do. But when this industry talks about itself, the words get in the way. To understand any industry, you have to understand what its people are talking about. Nobody in resources would expect a newcomer to know what JORC means, yet AI announcements assume the whole world speaks their language. What settled it for me was a recent presentation by Kimi, one of the Chinese AI companies now competing with the biggest American labs. I watched it wanting to learn, and the words were mind-numbing: token efficiency, optimisers, context length, agent swarms. If I was going to understand what these companies are building, and what the market is paying for, I had to decode the language first. This Insight is that decoding, for me as much as for you. The presentation was given by Yang Zhilin, the founder of Moonshot AI, the Beijing company behind the Kimi models, at NVIDIA's GTC conference in San Jose on 19 March 2026. The session was called "How We Scaled Kimi K2.5". Here is a taste of it, quoted as spoken: If that reads like another language, this piece is for you. By the end, you will be able to re-read that paragraph and follow every word. And because one company from Beijing is not the whole industry, each section pairs the Kimi talk with a second voice, OpenAI, Anthropic, Google, Microsoft, NVIDIA, Meta or xAI, saying the same kind of thing in their own words. The jargon is not one company's habit. It is the working language of a global industry that now carries trillion-dollar price tags. 2.00 — THE UMBRELLA WORDS: AI, MACHINE LEARNING AND WHAT AN LLM ACTUALLY IS Five nested words, and the engine behind the app Most confusion starts above the jargon, with words that sound interchangeable but are not. They nest inside each other like Russian dolls, and FIG. 01 draws the nesting. Artificial intelligence (AI) is the whole field: any attempt to make computers do things that normally need human intelligence. Machine learning is the approach that now dominates the field: instead of programmers writing rules, the computer works out its own rules from examples. Deep learning is machine learning done with a neural network, a program loosely inspired by the brain, built from many layers of simple connected units. Generative AI is the branch that produces new material, text, images, video and computer code, rather than just sorting or scoring things. And a large language model (LLM) is the kind of generative AI behind the chat tools: a very large neural network trained on enormous amounts of text to predict what comes next. The letters in GPT describe exactly that: Generative, Pre-trained, Transformer, where the transformer is the 2017 model design that made today's AI work at scale. You do not need to know how a transformer works, only that when you read the word, it means the standard engine design the whole industry builds on. The other distinction worth having is between the engine, the product and the company. The model is the engine. The chatbot or app is the car built around it. The company is the maker. OpenAI makes the GPT models that power ChatGPT. Anthropic makes Claude. Google makes Gemini. xAI makes Grok. Meta built Llama. Moonshot AI makes Kimi. When a headline says "GPT-5.6 beats Gemini", it is comparing engines, not apps, and certainly not businesses. 3.00 — THE TOKEN: THE UNIT EVERYTHING ELSE IS PRICED IN A small piece of a word, and the industry's whole accounting system Now the word in this article's title. A token is a small chunk of text, roughly three-quarters of an English word. Before an AI model reads anything, the text is chopped into these chunks, a step called tokenisation, and everything the model reads and writes is counted in them. OpenAI's own documentation gives the rule of thumb: 1,000 tokens is about 750 words. FIG. 02 shows a sentence cut into tokens; FIG. 03 puts sizes in scale. Why does one small word carry so much weight? Because the token plays three roles at once, and each role turns up in a different kind of announcement. The price role first. FIG. 04 compares what the major labs charged per million tokens in late July 2026. Two things stand out. Output tokens cost more than input tokens everywhere, because reading is done in one parallel pass while writing happens one token at a time and takes more computing effort. And the spread is enormous: from about US$0.28 per million output tokens at the cheapest Chinese model to US$90 at OpenAI's dearest professional tier, a gap of roughly 320 times. Price is a choice about quality, speed and brand, not a fixed cost of the technology. Then the usage role, which produced the most quoted number of 2026. Google's chief executive Sundar Pichai told the company's I/O conference in May that Google's products went from processing 9.7 trillion tokens a month two years ago to about 480 trillion a year ago, to "over 3.2 quadrillion per month" now, with its developer services handling "roughly 19 billion tokens per minute". FIG. 05 draws that curve. NVIDIA's founder Jensen Huang put the same idea in one sentence at GTC, the same conference where the Kimi talk was given: "Tokens are the building blocks of AI... your data center is now a factory to generate tokens." 4.00 — TRAINING AND INFERENCE: BUILDING THE PLANT AND RUNNING IT The two lives of a model, and the words that live inside each Every AI model has two lives. Training is the first: the model is shown trillions of tokens of text and slowly adjusts itself until it can predict language well. It happens once, takes months, and costs an enormous amount. Inference is the second: the finished model answers questions, every day, for everyone. It is the recurring cost of every token sold. FIG. 06 draws the split, in a shape any resources investor will recognise. Inside the model sit its parameters, also called weights: the billions of numbers the model adjusted during training, which hold everything it learned. Model sizes are quoted in parameters the way deposits are quoted in tonnes. Kimi K2, the model in our presentation, has one trillion of them. During training, progress is tracked by a single error score called the loss: the lower the loss, the better the model predicts. The arithmetic runs on GPUs, graphics processors from NVIDIA and others that turned out to be ideal for AI, or on TPUs, the equivalent chips Google designs for itself. And then there is the word that made me stop the Kimi video: the optimiser. The optimiser is the recipe the model follows while it learns, the set of rules deciding how much to adjust after each mistake. Almost the whole industry uses a 2014-era recipe called Adam. Moonshot's claim at GTC was that its new optimiser, called Muon (and MuonClip once stabilised), gets roughly twice the learning out of the same data. That is why Yang Zhilin said the line quoted in Section 1.00: with a two-times token efficiency, "50 trillion tokens" of data behaves like 100 trillion. Not magic; a better recipe. The claim is the vendor's own, from its own paper, and is best read that way. 5.00 — SCALING LAWS, TOKEN EFFICIENCY AND THE DATA WALL Why "bigger" has rules, and why the data is running out The AI boom rests on an empirical observation called the scaling laws: make the model bigger, feed it more data and give it more computing power, all together, and its results improve along a curve you can predict in advance. The landmark papers behind this were published by OpenAI researchers in 2020 and DeepMind researchers in 2022, and the industry has bet hundreds of billions of dollars that the curve keeps holding. OpenAI's Sam Altman stated it plainly in a 2025 essay: "The intelligence of an AI model roughly equals the log of the resources used to train and run it." But one of those three inputs is not like the others. Chips can be bought and models can be enlarged, yet the supply of good training text is finite. The industry calls this the data wall: the highest-quality human-written text, roughly speaking the useful part of the internet, has largely already been used. Anthropic's chief executive Dario Amodei put the underlying puzzle well in a February 2026 interview: "Humans don't see trillions of words. There is an actual sample efficiency difference here", meaning people learn language from far less data than models need. Two responses dominate. One is synthetic data: text generated by AI models to train other AI models. The other is token efficiency: squeezing more learning out of every token you already have, which is exactly what Moonshot's optimiser claim is about, and why a dry-sounding piece of mathematics was the headline of a major conference talk. 6.00 — CONTEXT WINDOWS: HOW MUCH THE MODEL CAN HOLD IN ITS HEAD The working memory, measured in tokens The context window is the model's working memory: the maximum number of tokens it can consider at once, covering both what you give it and what it writes back. Once a conversation or document exceeds the window, the earliest material falls out of view. Window sizes have become a headline specification, and FIG. 07 compares the current flagships. Why the arms race for bigger windows? Because the window sets what kind of work the model can take on. A model that holds a paragraph can answer a question. A model that holds a data room can audit it. In the Kimi talk, this is the second of the three things Moonshot says it scaled: their Kimi Linear design mixes a cheaper form of attention (the mechanism the model uses to weigh which earlier tokens matter) with the standard, expensive kind, in a three-to-one ratio, precisely to make very long working memories affordable. The stated goal: an AI that can run "for days or even weeks" on a single task without losing the thread. Anthropic's marketing makes the same promise from the other side of the Pacific: its Claude Sonnet 4.5 launch boasted that the model "maintains focus for more than 30 hours on complex, multi-step tasks". 7.00 — GETTING ANSWERS OUT: PROMPTS, REASONING, MULTIMODAL AND LOOKING THINGS UP The words you meet when you actually use the tools The vocabulary so far describes how models are built. This section covers the words that describe using them, the ones an investor meets in product launches rather than research papers. A prompt is simply the instruction you type, and prompt engineering is the craft of writing instructions that get better results. A reasoning model is the industry's newest product category: a model that can spend extra time working through a hard problem step by step before answering, producing an internal chain of thought. The extra thinking happens at answer time, so the industry calls it test-time compute, and the crucial commercial detail is that the thinking is billed in tokens like everything else. NVIDIA's Jensen Huang explained it at CES in January 2026 in one line: "We also have test-time scaling, which is another way of saying thinking. You think in real time." And in another: "Instead of a one-shot answer, inference is now a thinking process." Multimodal means one model handling more than one kind of material: text, images, audio and video together. The Kimi presentation is a worked example: Moonshot says K2.5 was the first open model trained on vision and text jointly "from day one" rather than bolting image skills onto a finished text model, and reported a result worth noticing: training the model on vision tasks alone improved its text results too, and a strong text model lifted its vision results. The claim is the vendor's own, but the direction, single models that see and read, is industry-wide. Then there is the machinery for making a model useful on your information. RAG, retrieval-augmented generation, lets a model look answers up in a nominated set of documents instead of relying on what it absorbed in training, which sharply reduces invention. The look-up usually runs on a vector database, a special store that searches by meaning rather than by exact words. Samso has covered vector databases before, in the AI & Capital piece "How to buy the AI build-out on the ASX", where our research found there is no pure-play way to buy one on the ASX. Guardrails are the rules wrapped around a model to stop harmful or off-limits outputs. Which brings us to the industry's most famous word for failure. A hallucination is a confident, fluent, wrong answer. It happens because a language model is a prediction machine: it produces the most plausible next tokens, and plausible is not the same as true. The rate has fallen with each generation, but it has not reached zero, and no vendor claims it has. 8.00 — AGENTS AND AGENT SWARMS: FROM ANSWERING TO DOING The industry's favourite word of 2026 An agent is an AI model given tools and permission to take steps on its own: search the web, run computer code, file the result, then decide what to do next. The adjective agentic now appears in nearly every AI announcement, and it marks the industry's core shift: from software that answers questions to software that completes tasks. The Kimi presentation's third big idea takes this one step further. In an agent swarm, an orchestrator (a lead agent) breaks a large job into pieces and hands them to a team of sub-agents working in parallel, then collects and assembles the results. Yang Zhilin's own analogy was a company: a chief executive decomposing work across departments, with researchers, developers and fact-checkers each doing their part. FIG. 08 redraws the structure he described. Moonshot's stated ambition is swarms of hundreds or a thousand agents finishing tasks fast enough "to produce real economical value". Every major lab is saying a version of the same thing. Google's Sundar Pichai told I/O 2026 "we're firmly in our agentic Gemini era". Microsoft's Satya Nadella told Build 2026 that "agents effectively are a new execution environment... a new paradigm", and described a security system where "over 100 specialized agents are working together to discover, debate and prove exploitable vulnerabilities end to end". Anthropic's Claude launches lead on "long-running agents" that keep checking their own work. When you read "agentic" in an announcement, this is the whole of what is meant: the software acts, in steps, with tools, increasingly in teams. 9.00 — THE WORDS OF THE RACE: OPEN WEIGHTS, FRONTIER, AGI AND THE REST The vocabulary of ambition, and what it is worth The last group of words describes not the machinery but the race itself, and these are the words that move markets. Open weights versus closed is the industry's deepest divide. An open-weight model's parameters can be downloaded and run by anyone, free: this is Kimi, DeepSeek and Alibaba's Qwen, and historically Meta's Llama. A closed or proprietary model can only be used through the maker's paid service: OpenAI's, Anthropic's and Google's flagships. Note what "open" does not include: usually not the training data or the full recipe, so open weights is not quite "open source" in the traditional software sense. The Kimi talk's framing, that open models are "closing the gap" with proprietary ones, is now conventional wisdom even at NVIDIA, whose chief executive told Axios in July 2026: "These Chinese models are excellent... Free AI should be great for chips." That the divide is strategy rather than ideology was proven in April 2026, when Meta, the company that built its AI reputation on giving Llama away, launched the first model from its Superintelligence Labs, Muse Spark, as a closed product, saying only that "we hope to open-source future versions of the model". The open champion went closed the moment it believed it had a lead; the challengers stay open because free distribution is how a challenger wins users from an incumbent. Then the ambition words. A frontier model is simply one of the handful of most capable models in the world at a given moment. AGI, artificial general intelligence, means an AI able to match humans across essentially all thinking work, and superintelligence means exceeding us. These are the words the money is raised on, and the industry's leaders now use them as near-term forecasts, on the record: Mark Zuckerberg in July 2025: "Developing superintelligence is now in sight." Sam Altman in June 2025: "Humanity is close to building digital superintelligence." Google DeepMind's Demis Hassabis on AGI, in 2026: "2030 is when I expect it to arrive, either plus or minus a year." Elon Musk went furthest in a January 2026 podcast, saying he expected AGI "next year". Alignment and safety name the work of keeping such systems doing what we intend; Anthropic in particular builds its brand on it. Three more words complete the set, all live in 2026. Distillation is training a cheap model on an expensive model's answers, the student copying the master's homework; in July 2026, Washington accused Moonshot of doing exactly this to American models, an accusation that is reported and denied, not proven. Sovereign AI is a country deciding it needs its own models, data and data centres rather than renting someone else's, the phrase behind national AI programmes worldwide, including the partnership between OpenAI and Australia's NextDC. And a wrapper is investor slang for a thin product built on someone else's model: worth remembering, because a company whose "AI product" is a wrapper owns a user interface, not the technology. 10.00 — FIVE QUESTIONS TO ASK OF ANY AI ANNOUNCEMENT The vocabulary, put to work Learning the words is only useful if it changes how you read. Here are five questions the vocabulary now lets you ask of any AI announcement, product launch or "AI strategy" slide. What is the unit? Tokens, parameters, users, gigawatts or dollars? Each unit flatters a different thing. A company quoting tokens is bragging about volume; parameters, about size; neither is revenue. Is that a usage number or a money number? "3.2 quadrillion tokens a month" measures work done, not dollars earned, and says nothing at all about profit. Who measured it? Vendor benchmarks are chosen by the vendor, from a rotating menu, after the fact. "Best on X" means best on the test they picked. Independent verification is rare and valuable. Which life of the model is the money for? Training (the build) or inference (the running)? Capital raised "for compute" can mean either, and they have completely different payback profiles. Who pays whom? Agents consume tokens; token sellers promote agents. Chipmakers invest in labs that buy chips. Follow the direction of the enthusiasm and check whether it matches the direction of the money. And now, the test. Here is the passage from Section 1.00 again: "Suppose you have 50 trillion high-quality tokens, and then you apply this new optimizer... all of a sudden you have a two times token efficiency... We introduce this new learning paradigm of agent swarms... at the end of the day, we're going to have a swarm of agents that each of them have a super long context." Read it once more. A company with a fixed supply of training text says its new learning recipe doubles what that text teaches its model; and its software now works as coordinated teams, each member able to hold enormous amounts of material in working memory. That is all it ever said. The ideas are ambitious; the language was the only barrier. The condensed decoder below covers the terms this piece taught. The full glossary, more than 45 terms in plain English, lives permanently on the Samso website, with the downloadable Samso AI Decoder PDF, and every future Samso piece on AI links back to it. References & sources This piece is built on one primary transcript (supplied) and the public statements, documents and pricing pages listed below. Quotes are reproduced verbatim from official transcripts or company pages except where noted as reported through a third-party transcript. Market-sensitive and dated figures (prices, context windows, token volumes) are quoted as at the dates shown and must be refreshed on publication day. No locator map appears because the subject is not geographic (house rule 4.2, waived). All figures are original Samso illustrations. Yang Zhilin (founder, Moonshot AI) — "How We Scaled Kimi K2.5", NVIDIA GTC, San Jose, 19 March 2026. Transcript supplied by Samso; session corroborated by Constellation Research, China Biz Insider and 36Kr coverage. Source of: the opening quotation; MuonClip optimiser and the two-times token efficiency claim; Kimi K2's one trillion parameters; Kimi Linear and the three-to-one attention mix; the agent swarm design and company analogy; K2.5's joint vision-and-text training claims. All capability claims are the company's own account. Jensen Huang (NVIDIA) — GTC 2026 keynote, 16 March 2026, official NVIDIA transcript ("Tokens are the building blocks of AI"; "your data center... is now a factory to generate tokens"; "three scaling laws"); CES 2026 keynote, January 2026, Rev.com transcript ("test-time scaling... another way of saying thinking"; "inference is now a thinking process"); interview with Axios, 22 July 2026 ("These Chinese models are excellent"; "Free AI should be great for chips"). Sundar Pichai (Google) — I/O 2026 opening keynote, 19 May 2026, official written version at blog.google. Source of: 9.7 trillion / ~480 trillion / ~3.2 quadrillion monthly tokens; "roughly 19 billion tokens per minute"; "we're firmly in our agentic Gemini era". Satya Nadella (Microsoft) — Build 2026 opening keynote, 2 June 2026, official Microsoft transcript. Source of: "tokens per dollar per watt"; "agents effectively are a new execution environment"; the 100-plus-agent security example. Sam Altman (OpenAI) — essays at blog.samaltman.com: "Three Observations" (February 2025; the log-of-resources scaling line), "The Gentle Singularity" (June 2025; "Humanity is close to building digital superintelligence"). Dario Amodei (Anthropic) — interview on the Dwarkesh Podcast, February 2026, quoted from the published transcript ("Humans don't see trillions of words..."); Anthropic, "Claude Sonnet 4.5" launch post, 29 September 2025 ("maintains focus for more than 30 hours"). Mark Zuckerberg / Meta — "Personal Superintelligence" letter, meta.com, 30 July 2025 ("Developing superintelligence is now in sight"); "Introducing Muse Spark", about.fb.com, 8 April 2026 ("we hope to open-source future versions of the model"). Demis Hassabis (Google DeepMind) — interview with Fast Company, 22 May 2026 (AGI "2030... plus or minus a year"); Yann LeCun — Davos panel remarks as reported by Fortune, 23 January 2026 ("completely LLM-pilled"; "digging the same trench"). Elon Musk (xAI) — post on X, 17 January 2026 (Colossus 2 "First Gigawatt training cluster in the world"); Moonshots podcast, January 2026, quoted from the published transcript (AGI "next year"). Podcast-transcript quotes to be re-verified against audio before publication. Pricing and specifications — official pricing/documentation pages of OpenAI, Anthropic, Google, xAI, Moonshot AI and DeepSeek, all fetched 26 July 2026. Known conflicts noted in the FIG. 04 and FIG. 07 captions (Claude context-window tiers; OpenAI's newest window unconfirmed on an official page; DeepSeek's launch-versus-current prices). Refresh all on publication day. OpenAI Help Center — "What are tokens and how to count them" (the 1,000-tokens-per-750-words rule). Scaling-law literature — Kaplan et al., "Scaling Laws for Neural Language Models" (2020); Hoffmann et al. ("Chinchilla"), "Training Compute-Optimal Large Language Models" (2022). Moonshot distillation accusation — reported by US press in July 2026 following the Kimi K3 release; an accusation, denied and unproven, and described as such in Section 9.00. Photographs (PLATE 1 to PLATE 5) — Unsplash, used under the Unsplash licence: Maximus Beaumont (taxi meter); Taylor Vick (data centre); Vishnu Mohanan (circuit board); Susan Q Yin (library); Frames For Your Heart (construction cranes). All retrieved 27 July 2026. Samso archive — "Tokens Are the Bill, Not the Value" (AI & Capital, Part 1, published 25 July 2026; source of the volume-versus-value framing revisited in Sections 3.00 and 10.00, and first publication of the Google token-volume figures redrawn here as FIG. 05) and "How to buy the AI build-out on the ASX" (AI & Capital, Part 2; source of the ASX vector-database finding cited in Section 7.00; link to be added on publication). Cover photo: Black Computer keyboard, Samiknoov. CC0, via Wikimedia Commons (https://commons.wikimedia.org/wiki/File:Black_Computer_keyboard.jpg). Reproduced photograph, shown faded under a navy overlay on the article thumbnail. The Samso Way – Seek the Research Here at Samso, we pride ourselves on delivering content for investors that is independent and informed by over three decades of experience in the industry. Our content is well-researched and is only created if I see merit in discussing the company's story. Our mission is simple: cut through the noise and spotlight what matters—genuine stories, grounded insights, and real opportunity. Our content is well-researched and is only created if the team sees merit in discussing the company or concept. Investors can explore our three core platforms: Coffee with Samso Samso Insights Samso News There may be numerous paths to success in investing, but the common thread among successful individuals is that they remain committed to making informed decisions. Equip yourself with the right knowledge and tools, and you will be well on your way to achieving your financial goals. Most importantly, investors need to be absolutely diligent in understanding their own risk-reward tolerance and capabilities. Never bite off more than you can chew. As they say, Rome wasn't built in a day, and the Great Wall stood because it took centuries to complete. The Samso Philosophy: Stay curious. Stay sharp. And remember—digging deeper always uncovers the real value. In Life, there is no such thing as a Free Lunch. Never bite off more than you can chew is my parting comment. Happy Investing, and the only four-letter word you need to know is DYOR. To support our independent work, please head over to our Support Page and give us a helping hand in any of the ways listed. This is a new initiative for the Samso Platform, and it was always the concept of Samso when we started this journey in 2018. Disclaimer The information or opinions provided herein do not constitute investment advice, an offer, or solicitation to subscribe for, purchase, or sell the investment product(s) mentioned herein. It does not take into consideration, nor have any regard to your specific investment objectives, financial situation, risk profile, tax position and particular, or unique needs and constraints. Read full Disclaimer. About Samso Samso is a trusted platform that equips dedicated investors with up-to-date industry knowledge and insights from top CEOs and thought leaders. By staying informed on business advancements and market trends, investors can enhance their financial decisions through a combination of expert guidance and their own research. Samso Insights | www.samso.com.au | An Investor Lens on ASX-Listed Companies