Extreme heat pushes electric demand on the mid-Atlantic grid, forcing Virginia’s data-center boom to rely on backup diesel generators and prompting emissions and health concerns as regulators weigh using standby power to prevent outages.
A study by researchers from UC Riverside and Caltech highlights the significant air pollution and health impacts associated with AI-related electricity consumption, predicting that by 2030, AI could cause 1,300 additional premature deaths annually in the U.S. due to increased emissions from data centers and diesel generators. The study estimates AI's public-health costs could reach $20 billion, with emissions potentially rivaling those from California's cars. The researchers call for greater transparency from tech companies regarding the air-pollution impacts of their AI operations.