AI applications in oil and gas exploration and production lower drilling costs, discover new fields, and improve recovery rates, resulting in cheaper and more abundant fossil fuels globally [1, 2, 3]. A peer-reviewed study published on August 10 in Nature journal reports these AI-driven productivity gains increase carbon emissions enough to offset reductions achieved by renewable energy technologies [1, 2, 4, 3].
Researchers from the nonprofit Enabled Emissions Campaign, founded in 2024 by former Microsoft sustainability leaders Holly and Will Alpine after they left over concerns of Microsoft’s fossil fuel involvement, highlighted that the emissions fueled by AI-enabled fossil fuel production could add between 500 million and 1.8 billion metric tons of CO2 annually if AI were applied across energy sectors [1, 3, 2, 4]. Holly Alpine said, "Most assessments of AI's climate impact are framed as a tradeoff between data center energy use and the emissions AI might help avoid through renewables and efficiency gains. What's missing from that calculation entirely is the other side of the same ledger: the emissions enabled from using AI to make fossil fuel production cheaper and more profitable" [2].
The study finds that for every 1% increase in fossil fuel productivity powered by AI, renewable energy output must increase 4% to 5% to maintain net-zero emissions targets [1, 3]. Will Alpine emphasized the need to consider both sides, saying, "One of the key insights of our paper is that you cannot treat them independently. They are two sides of the same coin" [4].
While current discussions focus heavily on the energy demand and emissions from AI data centers, the researchers estimate that AI-driven fossil fuel emissions exceed those from data centers by a factor of 2.8 to 10 times [1, 3, 2]. Meanwhile, the United States is expanding natural gas power plant capacity nearly threefold by 2025, partly to meet rising electricity demand from data centers [1, 3]. The US government has also extended the operational life of six coal-fired power plants originally scheduled for retirement [1, 3].
The American Petroleum Institute (API) disputes the notion that increased energy production necessarily conflicts with emissions cuts, citing ongoing investment in technology and improved operational practices. API spokesperson Andrea Woods said, "The U.S. oil and natural gas industry is continuing to produce more energy while reducing emissions by investing in better technology, implementing stronger operational practices and supporting science-based policy" [2].
Holly Alpine noted that sustainability efforts within technology companies often concentrate on direct operational emissions rather than the broader emissions enabled by AI applications in fossil fuel industries [4].
Multiple media outlets published reports on August 11 highlighting the new study and its findings about AI’s complex environmental effects [1, 2, 4, 3].
