Botswana Minerals Plc · LSE:BMIN
Botswana Minerals — Botswana Diamonds — African Mining Summit (AMS 2025)
34 slides from Botswana Minerals's conference presentation, Oct 15, 2025
Slide 1: An Artificial Intelligence-driven Approach to Mineral Exploration in Botswana
An Artificial Intelligence-driven Approach to Mineral Exploration in Botswana James AH Campbell Managing Director, Botswana Diamonds plc African Mining Summit (AMS 2025) Gaborone 15-16 October 2025
Slide 2: Agenda
African Mining Summit (AMS) 2025 An Artificial Intelligence-driven Approach to Mineral Exploration in Botswana Marsfontein blow pit and dumps when active AGENDA ⧫ Global demand: exploration crisis ⧫ AI foundations: remote sensing, geophysics, data fusion ⧫ Company landscape and case studies ⧫ Cost/t
Slide 3: Global demand: exploration crisis
Global demand: exploration crisis AMS 2025 3
Slide 4: Global Demand for Critical Minerals
Global Demand for Critical Minerals ⧫ Energy transition drives unprecedented demand for Cu, Li, Co, Ni, REEs. ⧫ Lithium demand projected >5× by 2040; Ni/Co/REEs to at least double. ⧫ Botswana prospective for Cu– Ni–Co–PGEs alongside diamonds. ⧫ Forecast supply gaps of 30–40% by mid-2030s. Deloitte,
Slide 5: Exploration Crisis: Rising Costs, Fewer Discoveries
Exploration Crisis: Rising Costs, Fewer Discoveries ⧫ Global discovery rates have gone down 75% since 2010 despite higher spending. ⧫ Greenfield success rate ~0.02% (1 in 5,000 drillholes). ⧫ Deposits under cover need new geoscience and computational tools. ⧫ AI offers systematic, data-driven target
Slide 6: Juniors vs Majors in Discovery
Juniors vs Majors in Discovery ⧫ >60% of discoveries in the past 20 years by juniors. ⧫ Majors focus on acquisitions; juniors chiefly underfunded with >90% failure rate. ⧫ If the junior is incubated, the failure rate is considerably lower. ⧫ AI reduces risk and improves financing potential for junio
Slide 7: AI foundations: remote sensing, geophysics, data fusion
AI foundations: remote sensing, geophysics, data fusion AMS 2025 7
Slide 8: Artificial Intelligence in Mineral Exploration
Artificial Intelligence in Mineral Exploration ⧫ Integrates geology, geochemistry, geophysics, and remote sensing. ⧫ Techniques include Machine Learning, ensembles, and Bayesian frameworks. ⧫ Outputs: prospectivity maps, ranked targets with uncertainty. ⧫ Continuous retraining improves decision qual
Slide 9: Remote Sensing and Geophysical AI
Remote Sensing and Geophysical AI ⧫ Hyperspectral/UAV imagery mapped with Machine Learning. ⧫ AI inversion of gravity, magnetics, and EM improves subsurface imaging. ⧫ Data fusion across methods reduces ambiguity. ⧫ Rapid screening focuses budgets on highest priority targets. Javan et al, 2024 AMS 2
Slide 10: Data Fusion and Prospectivity Modelling
Data Fusion and Prospectivity Modelling ⧫ Combines knowledge-driven models with Machine Learning. ⧫ Features: litho-structural context, basin architecture. ⧫ Ensembles yield class probabilities with uncertainty. ⧫ Governance avoids over-fitting and bias. Bell Geospace, 2024 AMS 2025 10
Slide 11: Company landscape and case studies
Company landscape and case studies AMS 2025 11
Slide 12: Select Companies Applying AI
Select Companies Applying AI ⧫ KoBold Metals: decision theoretical targeting. ⧫ Earth AI: vertical integration, higher hit rates. ⧫ Planetary AI: Xplore method for transparent prospectivity. ⧫ Specialists: GoldSpot, Mira (geophysics, hyperspectral). Mining Doc, 2024 AMS 2025 12
Slide 13: Case Study: KoBold Metals (Zambia Copper)
Case Study: KoBold Metals (Zambia Copper) ⧫ Mingomba discovery: largest Cu find in Zambia in a century. ⧫ AI targeting confirmed in ~1 year of drilling. ⧫ Portfolio learning enables redeployment across districts. ⧫ Algorithms prioritise drilling with maximum information. ⧫ Funded by Gates/Bezos with
Slide 14: Case Study: Earth AI (Australia)
Case Study: Earth AI (Australia) ⧫ ~75% drill-hole success on AI-generated targets. ⧫ Satellite analytics + geological Machine Learning + rapid drilling. ⧫ Cost reductions of ~80% via tighter targeting. ⧫ Data + execution integration improves confidence. Mining.com, 2025 AMS 2025 14
Slide 15: Planetary AI 'Xplore' (Botswana)
Planetary AI 'Xplore’ (Botswana) ⧫ 57+ deposit models + Machine Learning at basin/country scales. ⧫ Semantic AI grounds Machine Learning in geological context: boosting accuracy, clarifying uncertainty and elevating confidence. ⧫ Transparent prospectivity maps with quantified uncertainty. ⧫ Supports
Slide 16: AI Cost and Time Benefits
AI Cost and Time Benefits ⧫ AI narrows search space, improves hit ratios. ⧫ Exploration cycles compress with automation (4x faster). ⧫ Discovery cost reduction up to ~80%. ⧫ Industry savings: $300B+ annually by 2035. CleanTech, 2025 AMS 2025 16
Slide 17: Differential Impact: Juniors and Majors
Differential Impact Juniors Majors ⧫ AI processes legacy datasets ⧫ Portfolio-scale ranking and cheaply. optimisation. ⧫ Raises confidence, improves ⧫ Standardised data pipelines access to capital. enable comparability. ⧫ Outputs are transparent and ⧫ Decision analytics quantify ESG explainable. and
Slide 18: BOD strategy: dataset and AI-screened targets
BOD strategy: dataset and AI-screened targets AMS 2025 18
Slide 19: Botswana Diamonds plc (BOD): Strategy Overview
Botswana Diamonds plc (BOD): Strategy Overview ⧫ BOD applies AI to a national-scale archive. ⧫ Partnership with Planetary AI for ground selection. ⧫ Focus on polymetals and diamonds. ⧫ Objective: secure licences, drill-ready targets, JVs. AMS 2025 19
Slide 20: BOD Legacy Dataset (Scale & Composition)
BOD Legacy Dataset (Scale & Composition) ⧫ Covers ~225,000 km², 383 GB data: ⧫ ~375,000 line km airborne geophysics. ⧫ 606 ground geophysical surveys ⧫ ~358,000 soil sample results. ⧫ ~32,000 drill logs. ⧫ Data standardised and fused for AI. ⧫ Coverage: Ngamiland, SE Botswana, kimberlite terranes. ⧫
Slide 21: BOD Licence Applications: AI-Screened Results
BOD Licence Applications: AI-Screened Results ⧫ Applications over ~7,500 km². ⧫ 11 polymetallic + 3 kimberlite targets identified. ⧫ Selection based on permissive geology + adjacency. ⧫ Phased acquisition and drilling. AMS 2025 21
Slide 22: BOD's Competitive Advantage
BOD’s Competitive Advantage Five Pillar approach ⧫ Data. BOD already has an extensive database in Botswana and proposes to acquire more across Southern Africa. ⧫ AI technology. BOD also has access to this through an alliance with Planetary AI Xplore, which utilises a proprietary platform which unloc
Slide 23: Polymetallic and kimberlite examples
Polymetallic and kimberlite examples AMS 2025 23
Slide 24: BOD Polymetallic Targeting (Example)
BOD Polymetallic Targeting (Example) ⧫ Ngamiland: Cu-Pb-Zn-Ag ± V, Co, PGEs. ⧫ SE Botswana: Cu-Ni-Co with ultramafic/mafic hosts. ⧫ Historic Zn-Pb-Ag resources validate predictions. ⧫ Next: geochem/EM to refine drill collars. AMS 2025 24
Slide 25: BOD Kimberlite Targeting (Example)
BOD Kimberlite Targeting (Example) ⧫ KIM populations differentiated to guide dispersion. ⧫ Gravity lows + magnetics identify prospective zones. ⧫ Clusters distal from known fields suggest new pipes. ⧫ Captured in licence applications for drilling. AMS 2025 25
Slide 26: AI vs traditional methods: future of AI
AI vs traditional methods: future of AI AMS 2025 26
Slide 27: AI vs Traditional Exploration
AI vs Traditional Exploration ⧫ Traditional: sequential activities, slow feedback. ⧫ AI: parallel hypotheses, probabilistic ranking. ⧫ Explainability supports disciplined budgets. ⧫ Higher conversion, lower discovery cost. RASonly, 2025 AMS 2025 27
Slide 28: Possible Future State (2030–2035)
Possible Future State (2030–2035) ⧫ District digital twins update in real time. ⧫ Autonomous UAVs/robotic drills with active learning. ⧫ Next-generation sensors probe deeper cover. ⧫ Geologists as data scientists. Yang et al, 2024 AMS 2025 28
Slide 29: Policy, Investment and ESG Context
Policy, Investment and ESG Context ⧫ Strategic finance favours efficient and successful commercial discovery. ⧫ AI reduces footprint by focusing on the best targets. ⧫ Auditable models aid permitting and engagement. ⧫ Botswana's stable jurisdiction enhances the case. EY.com, 2024 AMS 2025 29
Slide 30: Conclusions and Strategic Outlook
Conclusions and Strategic Outlook ⧫ AI reshaping discovery economics. ⧫ Botswana is an ideal testbed given the geology/data. ⧫ Juniors increase financing potential, and majors gain optimisation. ⧫ Early adopters will dominate the next cycle. Yang et al, 2024 AMS 2025 30
Slide 31: Thank You: Questions & Discussion
Thank You Questions & Discussion Photo: Andreas Stelzer AMS 2025 31
Slide 32: About the Author
About the Author ⧫ James Campbell is Managing Director of Botswana Diamonds plc (a diamond development company active in Botswana, South Africa and Zimbabwe and listed on London AIM and the Botswana Stock Exchange). He has spent over forty years in the diamond industry in a variety of leadership rol
Slide 33: References
References ⧫ International Energy Agency (2025). Global Critical Minerals Outlook. Paris: IEA. ⧫ Botswana Diamonds plc (2024–2025). Strategy and AI exploration materials. ⧫ Planetary AI. Xplore prospectivity platform overview and method notes. ⧫ KoBold Metals case reports and industry coverage (e.g.
Slide 34: Contact details
162 Clontarf Road Clontarf Dublin 3 Ph: +27 83 457 3724 Web: www.botswanadiamonds.co.uk Twitter: @BotswanaDiamond Email: james@botswanadiamonds.co.uk AMS 2025 34