The energy transition is a metals transition, decarbonisation technologies demand significant volumes of metals and minerals. Electric cars need 6x the mineral inputs of conventional cars. An offshore wind plant requires 13x more mineral resources than a similarly sized gas-fired plant.

As a result, demand for certain critical minerals is expected to skyrocket. Copper demand will 2x by 2050, and demand for Lithium will 15x according to some forecasts. 

To meet demand, new reserves of metals and minerals will need to be identified. Supply shortages are expected in lithium, copper, nickel and cobalt. Reserves might need to expand by up to 30% for copper and 70% for nickel, and to meet total expected demand between 2020–50. 

Finding new reserves has always been difficult and expensive, and it’s only getting more so. Most of the world’s easily discovered reserves are already being tapped. The ones that remain are in remote locations or deep underground. At the same time, the threshold for what miners define as an ‘economically viable’ reserve is only rising due to higher borrowing cost and, inflated operating costs.

Current approaches of exploration for new reserves are labour-intensive and success rates are extremely low. Mineral exploration today is largely manual. Geologists identify patterns in maps and data which is slow, inefficient, and prone to unpredictable discoveries. More than 99% of exploration projects fail to become mines, and the industry now spends three times more to make 60% fewer discoveries compared to 30 years ago.

ExplorationTech aims to make the exploration process faster, cheaper and more efficient by applying machine learning and artificial intelligence to exploration for reserves of critical minerals.

ExplorationTech startups build titanic databases of information about the Earth’s crust including geologic reports, soil samples, satellite imagery, and academic research papers. Kobold uses this data to train algorithms to identify patterns in the geology of places where metals were found in the past. Models that can only get better over time, as with other data network effects.

The algorithms can then be let loose on the full database of Earth’s crust data to generate a series of maps indicating where new reserves of critical minerals are likely to be found.  A software overlay over the physical world.

Supporting domestication of Supply: Current operational mines and refineries are concentrated in a small number of geographies, increasing risk and magnitude of operational disruption. Over 70% of lithium is mined in just three nations. Many nations are looking to mitigate these risks by building domestic supply. ExplorationTech can help build a pipeline of home-grown assets to shore up local supply

Technological advancement of Mining is lagging: Mining majors have underinvested in innovation, with slimmed down R&D programs since the 80’s. ExplorationTech companies have the opportunity to become the software arm of the early-stage mining industry. Several large mining companies have launched initiatives to invest in exploration technology, like BHP Xplor.

Increasing Scrutiny on Responsible Mining: Governments are increasingly scrutinising the environmental impact of mining operations. For example, In Chile, new regulations limit the number of miners that can commence exploration within the same area. This has sparked a race to gain rights for early-stage exploration.

Technological Effectiveness - Even with help from AI, placing bets on potential mineral deposits is far from foolproof. Metals often turn up in places with wildly different conditions and geologic histories. Training an algorithm on data from Alaska and applying it to Nevada means it might have a lot of wrong assumptions. Kobold’s discovery of a Zambia mine has been promising but physical discoveries are still limited.

Competition / defensibility - Larger technology providers such as IBM Watson are beginning to establish their own platforms as a service to mining companies. Mining companies themselves also have large in-house exploration teams progressively acquiring more data talent.

Timing the cycle - Mining is notoriously cyclical, with volatile commodity prices and investment patterns as a result. This can put pressure on investment into new technologies and exploration when outlook is poor. This poses a secondary risk for miners with skin in the game, it increases uncertainty of future demand/profitability of mining assets they develop.