AI Data Centers Fuel Pollution, Health Risks in the U.S.

The rapid growth of AI data centers is causing a significant increase in electricity demand across the United States, projected to rise from 4.4% of national consumption in 2023 to between 6.7% and 12% by 2028, and potentially 11 times higher by 2030. This surge is creating strain on existing power grids, prompting the consideration of keeping aging coal-fired power plants operational, as evidenced by a 2026 Trump administration decision to roll back federal soot standards citing AI energy needs. The environmental and health consequences are substantial. Researchers at the University of California, Riverside, project that increased particulate pollution from fossil fuel plants, exacerbated by data center energy demands, could lead to up to 1,300 premature deaths annually in the U.S. by 2030. Michael Cork of Harvard University's T.H. Chan School of Public Health is further quantifying these public health costs. Local and state governments are beginning to react to these impacts. St. Louis, Missouri residents are actively fighting particulate pollution, while in Virginia, AI data centers already consume 26% of the state's electricity, triggering over 60 legislative bills. In a notable local development, Los Angeles County supervisors are considering a moratorium on data center construction and a formal health impact assessment, reflecting a wider pattern of local governments confronting infrastructure decisions with limited community input.