As countries gather in Antalya, Turkey for the United Nations climate conference, artificial intelligence is expected to be a central topic. The host has urged nations to commit to designing, powering, and deploying AI in ways that support climate action. Yet despite optimistic claims from AI leaders, industries worldwide are doing little to use the technology to meaningfully cut the greenhouse-gas emissions driving global warming. OpenAI CEO Sam Altman has predicted that AI would unleash astounding triumphs, including solving the climate crisis, and a Google sustainability executive has argued the technology could sharply reduce global emissions. Such projections deserve scrutiny against present-day evidence.
A peer-reviewed study led by researchers at the London School of Economics Grantham Research Institute and the firm Systemiq estimated that AI could reduce global emissions by one to 5.3 billion tons of carbon dioxide annually, even after accounting for AI’s own energy use. The gains are concentrated in three sectors: power, meat and dairy, and vehicles. A follow-up review of public information from major players in each sector, conducted with AI assistance, found that few companies have announced pursuing the specific emissions-reducing approaches the study modeled, and fewer still report meaningful climate results.
The power sector shows the clearest potential. AI-driven forecasting of supply and demand could increase renewable electricity delivered from existing solar and wind plants by up to 10 percent, preventing an estimated one billion tons of emissions yearly. However, of the grid operators reviewed across China, the United States, and Europe, only one had deployed a proven, working AI program of this kind. A partnership between the UK’s National Grid ESO and Open Climate Fix halved solar forecasting error, avoiding hundreds of thousands of tons of carbon dioxide annually. Elsewhere, public evidence of such integration is scarce.
The meat and dairy sector presents the starkest gap between technical promise and market reality. Achieving the projected reductions would require AI to help alternative proteins match the taste and cost of conventional products. Some progress exists, such as a joint venture between the AI firm NotCo and Kraft Heinz. Still, the sector faces slow consumer adoption and falling investment, with venture funding for cultivated meat dropping sharply from its peak. Transportation offers more encouraging examples, including battery maker CATL’s AI system that reportedly sped materials screening by roughly a hundredfold and ride-sharing platform Didi Chuxing’s AI routing that lowered per-trip emissions. Experts caution, however, that oil and gas companies are also pursuing AI, potentially to expand fossil fuel production, and that integrating AI is only a first step toward real climate benefits.
Source: The climate case for AI has not been made via MIT Technology Review