Research quantifying artificial intelligence’s climate impact across the full power sector indicates that productivity gains enabled by the technology in fossil fuel operations exceed emissions reductions achieved through renewable energy applications. The analysis examined 64 different scenarios and found net yearly carbon pollution rose between 0.47 and 1.8 gigatonnes, representing approximately 1-5% of the energy sector’s annual emissions.
Previous studies have focused on indirect climate benefits of AI, such as reducing downtime at renewable facilities and optimizing electricity grids, while largely ignoring increased pollution from enhanced productivity in oil and gas extraction. According to researchers, net emissions only declined when scenarios excluded AI-driven productivity improvements in the fossil fuel sector. For emissions to reach equilibrium with similar AI adoption rates across both clean and dirty energy facilities, renewable productivity gains would need to outpace fossil fuel productivity gains by at least fourfold.
The International Energy Agency projects that AI could expand technically recoverable oil and gas reserves by 5% and reduce deepwater offshore project costs by 10%. Energy companies have already deployed AI at scale, with major operators reporting significant productivity increases and discovery improvements. Saudi Aramco reported embedding AI throughout operations to boost well counts, while Equinor attributed 27 recent discoveries on the Norwegian continental shelf to seismic technologies and AI. An independent energy research firm estimated that digitalization and AI would generate nearly $500 billion in cumulative value for fossil fuel companies between 2026 and 2030.
Researchers excluded datacentre energy demand from their calculations but found that AI-enabled productivity gains in fossil fuel operations result in emissions at least three times higher than current estimates for datacentre power consumption. The study noted results represented directional structural findings rather than precise forecasts, though the relationship persisted across all scenarios tested. Climate analysts highlighted concerns that unchecked AI industry expansion would inherently boost fossil fuel use beyond datacentre energy demands alone.
Article Attribution | Read More at Article Source
Article summary produced by Claude AI