AI may boost fossil fuel emissions more than clean energy gains

AI-driven productivity gains in fossil fuel production could increase climate pollution more than AI applications in renewables can reduce it, according to modelling of the power sector. Across 64 scenarios, researchers found net yearly carbon pollution rose by 0.47-1.8 gigatonnes, or about 1-5% of the energy sector’s annual emissions.

Net emissions fell only when AI did not improve productivity in coal, oil and gas. If fossil fuel and renewable energy facilities adopted AI at similar rates, renewables would need productivity gains at least four times greater than fossil fuels for emissions to break even. Researchers described the results as a directional and structural finding rather than a precise forecast.

The International Energy Agency estimates that AI could boost technically recoverable oil and gas reserves by 5% and cut the cost of deepwater offshore projects by 10%. Saudi Aramco has said it embedded AI “in everything,” while Equinor attributed 27 discoveries on the Norwegian continental shelf to new seismic technologies and AI.

Rystad Energy estimated that digitalisation and AI would create close to $500bn (£370bn) in cumulative value for fossil fuel exploration and production companies between 2026 and 2030 through more efficient operations, increased production, and shorter development timelines.