Data centers’ energy usage has increased due to the growing demand for artificial intelligence (AI), which presents serious health hazards to the general public. Emissions from U.S. data centers are expected to result in 1,300 premature deaths and 600,000 instances of asthma per year by 2030. With an estimated health cost of over $20 billion, these emissions, which are mostly from institutions that rely on fossil fuels, may be responsible for more than one-third of asthma-related fatalities in the country.
Research led by Shaolei Ren of the University of California, Riverside, highlights the widespread impact of data center pollution. “Public health impacts are direct and tangible,” Ren stated, noting that pollution travels far beyond the vicinity of data centers, affecting people nationwide. Fine particulate matter, a key pollutant, is particularly harmful, with the energy required to train a single AI model equivalent to emissions from 10,000 round trips between Los Angeles and New York by car.
The researchers used the U.S. Environmental Protection Agency’s tools to calculate these projections. By 2030, data centers could rival the U.S. steel industry in public health costs and match the emissions impact of vehicles in large states like California. Currently, Virginia’s Data Center Alley exemplifies this issue, with backup generators causing up to 14,000 asthma cases annually and costing $300 million in health impacts, even at minimal operation levels.
Many data centers still rely heavily on fossil fuels, even though some major corporations are investing in nuclear and renewable energy to reduce emissions. By 2030, data centers may use as much energy as California or New York if present trends continue.
University of Pennsylvania researcher Benjamin Lee urges more study to improve these approximations and investigate mitigating measures. Switching to cleaner energy sources will be necessary to meet this challenge and reduce the negative effects on the environment and human health.