Expert commentary
Joaquin Vespignani - Associate Professor of Economics, University of Tasmania, Australia
Joaquin Vespignani is an Associate Professor of Economics at Tasmanian School of Business and Economics, University of Tasmania; Program Director at the Centre for Applied Macroeconomic Analysis at the Australian National University; and research associate at the Globalization and Monetary Policy Institute at the Federal Reserve Bank of Dallas, US. Joaquin is also a Co-Founder of Commodia, a platform that connects academic economists with global energy markets.
Russell Smyth - Professor and Deputy Dean in the Monash Business School, Australia
Russell Smyth is Professor and Deputy Dean (Research) in the Monash Business School; an Elected Fellow of the Academy of Social Sciences of Australia and a Co-Editor of Energy Economics.
Case Study:
Artificial intelligence and critical minerals for energy transition
One of the most significant obstacles to the global energy transition toward decarbonization is the unprecedented amount of critical minerals required over the next two decades to support clean energy technologies. In the first four stages of the mining project development process (exploration, scoping, feasibility and development), these critical minerals often encounter both technical and non-technical barriers.
For example, AI improvements in the mining stage enhance ore grade control and optimize extraction processes, which directly increase productivity and reduce waste. Advances in AI have also facilitated secondary mining, which involves extracting valuable materials from previously processed waste or tailings. This approach not only addresses environmental concerns associated with mining waste, but also creates an additional source of critical minerals, thus, enhancing the overall efficiency and sustainability of mining operations (Maest, 2023)1.
In Figure 1, we show estimates of a benchmark production function to project how AI-driven productivity gains could increase lithium production in the mining stage. The general conclusions from these findings are applicable to most critical minerals. The grey line represents the International Energy Agency’s (IEA) (2024)2 Announced Pledges Scenario (APS), which projects that lithium production will increase from 0.2 Mt in 2023 to 1 Mt by 2040.
This contrasts with the IEA's Net Zero Scenario (NZS), which projects that 1.4 Mt of lithium will be needed by 2040 to realize carbon net zero targets. The light blue line represents a scenario in which AI leads to a moderate 1.5% per annum improvement in productivity. These projections indicate an increase of around 17% in lithium production, resulting in approximately 1.2 Mt by 2040. The green line represents a scenario in which AI results in a much larger improvement in productivity of 3% per annum, implying an increase in productivity of around 36% and total production of 1.39 Mt by 2040, just slightly below the NZS of 1.4 Mt projected by the IEA.
Figure 1: Productivity growth in the production of lithium for moderate and high adoption of AI scenarios, relative to the IEA's Announced Pledges Scenario over the period 2023 to 2040
These projections highlight the important role that AI can play in realizing the NZS by 2050. With the IEA (2024) estimating that $800 billion investment will be required in the mining stage, AI-driven productivity growth in mining can reduce the critical minerals shortfall.