eWEEK’s content and product recommendations are editorially independent. We may earn money when you click on links to our partners. Learn more.
A recent study by researchers at Caltech and the University of California, Riverside, highlights that the enormous energy demands of AI are creating dangerous air pollution, which could lead to 1,300 premature deaths in the United States each year by 2030, as well as pollution-related health costs reaching $20 billion per year.
Entitled “The unpaid price: quantifying the public health impact of AI”, this research article highlights the often underestimated ecological footprint of advances in AI. As more data centers and computing facilities emerge to power AI systems, energy consumption increases significantly. This increase in electricity demand leads to increased emissions from power plants and diesel backup generators, used to ensure a constant energy supply. These emissions contain fine particles, less than 2.5 micrometers, which penetrate the lungs.
Disproportionate impact on vulnerable communities
Researchers found that low-income communities, located near power plants and data centers, face the highest health risks. Diesel generators, often used as backup power sources, emit between 200 and 600 times more nitrogen dioxide than natural gas-fired plants, degrading air quality in these areas. However, the pollution does not stop there. It crosses state lines, affecting communities far from the original source.
“Pollution from backup generators in data centers in Northern Virginia migrates to Maryland, West Virginia, Pennsylvania, New York, New Jersey, Delaware and the District of Columbia,” the study says , “generating regional public health costs of $190 to $260 million per year. If these generators emit at their full potential, the annual cost could reach between $1.9 billion and $2.6 billion. »
The cost of AI-powered progress
The research paper highlights the incredible scale of this problem. Training a single large language model, such as Meta’s Llama 3.1, generates air pollution equivalent to 10,000 round-trip car trips between Los Angeles and New York. By 2030, AI-related pollution could rival the emissions from California’s 35 million vehicles, surpassing even the steel industry’s impact on public health.
Shaolei Ren, an associate professor at UC Riverside and co-author of the study, highlighted a significant gap in corporate responsibility.
-“When you look at sustainability reports from tech companies, they only focus on carbon emissions and water use,” Ren said. “There is absolutely no mention of harmful air pollutants, which already cause a public health burden. »
The authors encourage AI companies to adopt more transparent practices by reporting their carbon emissions and air pollution resulting from their activities. They also recommend compensating affected communities and adopting sustainable energy practices to reduce environmental damage.
“If you have family members with asthma or other health conditions,” Ren added, “air pollution from these data centers could be affecting them right now. This is a public health issue that we urgently need to address. »
Learn more about the energy consumption involved in powering AI and its impact on resources and the environment.
The question raised here calls into question the benefits of technological innovation in relation to its environmental and societal costs. How do we balance advances in AI with the need to protect the health of vulnerable communities? This complex subject requires continuous dialogue between sector players, decision-makers and citizens.
- --
