Californiaβs phase out would begin in 2026, when the state will require 35% of automakersβ sales to be zero emissions vehicles.
Β© 2024 TechCrunch. All rights reserved. For personal use only.
Nathan Howard/Getty Images
Researchers have found that the training of one large AI language model β like Meta's Llama 3.1 βΒ would generate as much air pollution as a car driving round-trip from New York to Los Angeles 10,000 times. The total cost of AI's health impacts, they found, could reach $20 billion within six years.
The team of researchers from the University of California at Riverside and the California Institute of Technology conducted what they say may be the first study of its kind assessing AI's impacts on air pollution. The paper, "The Unpaid Toll: Quantifying the Public Health impact of AI," which will be released later today, finds that the generation of electricity for data centers hosting artificial intelligence applications could pollute the air so much that by 2030 an additional 1,300 people may die prematurely each year as a result.
That would be a 36% increase over the current annual asthma-related deaths in the country.
The researchers β led by Shaolei Ren of UC Riverside and CalTech's Adam Wierman β examined the release of nitrogen dioxide, sulfur dioxide, and particulate matter with a diameter of 2.5 micrometers or less, which can penetrate deep into the lungs, by power plants and diesel generators associated with AI facilities.
The boom in artificial intelligence has resulted in a spike in electricity demand. McKinsey & Company, the consulting firm, projects that data centers will use 11 to 12% of the total electricity consumed in the United States in 2030, up from 3 to 4% last year. While the carbon emissions and water usage implications of that growth have started to draw scrutiny, the direct health impacts of the air pollution these facilities generate have been mostly ignored.
"There is something like this, air pollution, which is affecting people right now," Ren said in an interview. "We aren't paying attention to it at all."
The researchers estimate that the generation of electricity for AI data centers could trigger roughly 600,000 asthma symptom cases a year by 2030.
Last year, the researchers estimate, the generative AI boom led to a public health burden of $5.6 billion. By 2030, they calculated, AI's electricity-related public health costs will top $20 billion. That's more than double the public-health costs of coal-based US steelmaking, they write, and will rival the emissions produced by California's 35 million cars.
The cost estimates are based on a risk assessment tool developed by the Environmental Protection Agency that assigns a dollar figure to what it would take to avoid negative health outcomes, such as premature deaths, asthma symptoms, heart attacks, and missed days of school or work.
For Ren, whose field of study is responsible artificial intelligence, his interest in air quality dates back to his childhood. He lived in a small mining town in China until the age of six, where he saw a correlation between poor air quality and adverse health outcomes in his community, including lung cancer.
The paper is what's called a preprint, a standard practice in computer science research in which researchers make a paper public before submitting it for peer review.
The researchers also examined air pollution driven by emissions from diesel generators used by data centers for backup power and by the manufacturing of the silicone chips used in artificial intelligence.
To examine the impact of diesel-generators, the researchers looked at those permitted in Virginia, home to one of the densest collections of data centers in the world. Generators, according to the paper, produce 200 to 600 times the nitrogen dioxide per unit of power produced than a natural gas power plant.
"Diesel generators represent a major source of on-site air pollutants for data centers and pose a significant health risk to the public," the researchers wrote.
Even assuming that emissions by Virginia-permitted generators were just 10% of what the commonwealth's regulations allow, they would cause an additional 13 to 19 deaths each year. If the diesel generators emitted 100% of what is allowed, they would lead to 130 to 190 additional deaths, the researchers found.
The public health burden of Virginia's data center generators amounts to $220 million to $300 million a year under the 10% assumption and as much as $3 billion a year under the 100% assumption, according to the study.
The researchers found those health effects are not contained to the state, as air pollution travels. They found it's actually a county in Maryland β Montgomery County β that is most affected by Virginia's AI generators.
The effects, according to the EPA tool, would also be felt in West Virginia, New York, New Jersey, Pennsylvania, Delaware,Washington D.C., and as far away as Florida.
"We thought the air pollution was limited to a small area," Ren said. "That's not true. There is actually cross-state air pollution." He and the other researchers found that the harmful effects are felt disproportionately by "economically-disadvantaged communities."
Ren said that understanding the broad dispersion of negative health outcomes could encourage AI companies to alter their site locations or AI training schedules. Health impacts are higher during the day, he said, and some locations have higher or lower health effects.
The researchers also call for greater transparency from the big tech companies that lead large language model training.
Those companies, including Amazon, Google, Microsoft, and Meta don't currently detail the air pollution impacts of their AI operations in their annual sustainability reports, the researchers wrote.
"If you look at the sustainability reports from these companies, they mention carbon and water, but they don't mention anything about air pollution," Ren said. "They should start reporting this in the same way."
Shutterstock
Some auto executives see an opportunity with the incoming Trump administration to roll back state rules requiring a rapid uptick in electric vehicle sales.
Executives at Nissan, Toyota, and the auto industry's largest US lobbying group say it will be impossible for the industry to meet aggressive timelines to phase out gas-powered cars and trucks by 2035 in a dozen states including California and New York, as well as Washington, DC. In six states, a target kicks in in 2026, when at least 35% of new car sales must be EVs.
"It will take a miracle to be achieved," JΓ©rΓ©mie Papin, senior vice president of Nissan, said this week during an event in Washington, DC. "That's where others need to do a reality check on what's possible."
He noted that EVs accounted for about 9% of new car sales nationwide in the third quarter β a record, but still far short of what regulators are requiring by 2026.
Automakers, facing lower-than-expected demand for EVs this year, are pulling back on production, and some companies are cutting jobs to save costs. At the same time, they have poured billions of dollars into EVs and executives say they are committed to the transition, especially to stay competitive with China as it churns out more affordable EVs. That balancing act has put the industry in a delicate position with Trump who railed against EVs on the campaign trail, vowing to kill tax credits and other incentives encouraging Americans to buy them.
Now the industry is strategizing how to influence Trump, including on EV sales requirements they view as too ambitious. Trump will likely take their side.
At a campaign event in Michigan in October, he said no state would be allowed to ban gas-powered cars. Trump during his first term tried to revoke California's authority to set stricter limits on tailpipe pollution than the federal government. California is granted that authority under the Clean Air Act but must get waivers from the Environmental Protection Agency. Biden restored the states' authority β a move currently being litigated and could reach the Supreme Court.
To avoid uncertainty, a group of automakers, including BMW, Ford, Honda, and Volkswagen, struck an agreement with California in 2020 to follow the state's rules through 2026.
The rules are stricter than federal regulations issued earlier this year by the Biden administration's EPA. Those federal rules aren't an "EV mandate," as Trump falsely said on the campaign trail. Rather, automakers can choose how to curb greenhouse gas emissions from cars, trucks, SUVs, and vans sold between 2027 and 2032. The agency estimated the rules could boost EVs to up to 56% of new car sales, with the rest from a mix of hybrids and gas vehicles.
Dealers, which were the first to sound the alarm on changes in the EV market last year, have argued that state and federal emissions requirements are out of step with demand. As companies push to meet these requirements, dealers complain they are stuck with unpopular EVs on their lots.
"A majority of customers are simply not ready to go electric right now," Dave Kelleher, a Chrysler-Dodge-Jeep-Ram dealer in Pennsylvania, told BI. "Maybe with a new administration, some of those fines will become a thing of the past, or even mitigated."
Karoline Leavitt, spokeswoman for Trump's transition effort, said Trump will stop attacks on gas-powered cars.
"When he takes office, President Trump will support the auto industry, allowing space for both gas-powered cars AND electric vehicles," she said in an email.
John Bozella, president of the Alliance for Automotive Innovation, which represents companies producing nearly all the new vehicles sold in the US, sent a letter to Trump in November asking that he ease emissions regulations but keep EV tax incentives fueling domestic investment in the supply chain.
An analysis commissioned by the Natural Resources Defense Council found that companies have announced $312 billion in planned investments in EVs and battery production since Biden took office in 2021, fueled partly by tax incentives in the Inflation Reduction Act.
One automaker, Toyota, supports doing away with EV mandates and subsidies altogether. In a recent op-ed in the Wall Street Journal, Toyota Chief Operating Officer Jack Hollins wrote that state mandates distort the market because companies funnel zero-emissions vehicles to those locations and ultimately limit choices for customers.
General Motors initially sided with Trump in his crusade against California's EV rules, but dropped its support of the legal battle after Biden won the 2020 election. It's unclear whether the automaker would once again side with Trump if he tries to roll back emissions requirements. Paul Jacobson, General Motors' executive vice president and chief financial officer, told reporters that ideally there'd be more consistency between federal and state rules. But the automaker will respect regulators' authority, he said.
"There's a lot at stake here," Jacobson said during the event in Washington. "That's why we talk about being nimble across the board, because sometimes it's the marketplace and sometimes it might be the regulatory environment. But we can't make excuses for poor performance. It's not just Washington. It's China, it's Europe. There's a lot of things going on all over the world and we have to be able to respond to that."
