RNS: The growing demand for Artificial Intelligence (AI) and the accompanying increase in computer processing are escalating air pollution levels, primarily from power plants and backup diesel generators. Experts now warn that by 2030, this pollution could lead to 1,300 premature deaths annually in the United States.
A study conducted by scientists from UC Riverside and Caltech reveals that the public health costs associated with this pollution, including diseases like cancer and asthma, as well as lost work and school days, are approaching $20 billion per year. The study’s findings were published online this week.
Shaolei Ren, a UCR associate professor of electrical and computer engineering and a corresponding author of the study, noted that tech companies often focus only on carbon emissions in their sustainability reports, overlooking the harmful air pollutants from their energy consumption. These pollutants are already burdening public health, Ren emphasized.
The study’s authors, including Caltech’s Adam Wierman, call for new standards that would require tech companies to disclose the air pollution linked to their energy use and backup generators. They also suggest compensating communities most affected by pollution from data centers.
The research highlighted that low-income communities are disproportionately impacted by air pollution, due to their proximity to power plants or backup generators at data centers. Moreover, pollution from these facilities can drift across state lines, affecting broader regions.
For instance, air pollution from backup generators at data centers in Northern Virginia affects surrounding areas such as Maryland, West Virginia, Pennsylvania, New York, New Jersey, Delaware, and Washington, D.C., leading to regional public health costs of $190 million to $260 million annually. If emissions reach their maximum allowed levels, these costs could skyrocket to $1.9 billion to $2.6 billion per year.
In some cases, the public health costs of pollution from AI processing centers surpass the electricity costs paid by tech companies, the study found.
As AI technology continues to reshape various industries, it is driving a surge in energy demand, making data centers the fastest-growing sector for energy consumption. This rise in energy use is expected to significantly increase air pollution, with the resulting health impact by 2030 projected to surpass that of the U.S. steel industry and rival the emissions from all vehicles in California, according to the study.
Ren concluded, “The growth of AI is driving an enormous increase in demand for data centers and energy, making it the fastest-growing sector for energy consumption across all industries.”
