Artificial intelligence (AI) technology has been advancing quickly, and demand is growing across the world. With this shift, the need for electricity is also growing at a potentially unsustainable rate. Experts worry that power development will not be able to keep up with AI growth. AI can also fuel misinformation and harmful spending which can worsen the climate crisis. Can the negative side effects of AI coexist with the technology’s potentially positive attributes?
How does AI affect the climate?
AI is gradually becoming a larger part of our lives as it carves its way into a number of industries and everyday technologies. “The artificial intelligence compute coming online appears to be increasing by a factor of 10 every six months,” said Elon Musk at the Bosch Connected World conference. (AI “compute” refers to the “computational resources required for artificial intelligence systems to perform tasks, such as processing data, training machine learning models, and making predictions,” said Komprise.)
Questions are being raised as to how to accommodate the large amount of data required for AI. The raw materials required for its hardware need to be mined, which can be “really labor-intensive and also environmentally expensive,” Shaolei Ren, associate professor of electrical and computer engineering at the University of California, Riverside, said to CBS News.
The biggest risk AI poses for the climate comes from the sizable computing it requires. “It is important for us to recognize the CO2 emissions of some of these large AI systems,” Jesse Dodge, a research scientist at the Allen Institute for AI in Seattle, said to Scientific American. It is also difficult to ascertain how much AI will truly affect the climate because “different types of AI — such as a machine learning model that spots trends in research data, a vision program that helps self-driving cars avoid obstacles or a large language model (LLM) that enables a chatbot to converse — all require different quantities of computing power to train and run,” said Scientific American.
Computing requires electricity, and the power supply is more at risk than we think. “Across the board, we are seeing power companies say, ‘We don’t know if we can handle this; we have to audit our system; we’ve never dealt with this kind of influx before,'” said Andy Cvengros, managing director of data-center markets at JLL, to The Washington Post. “Everyone is now chasing power.” The AI boom may also slow down the push toward renewable energy, “as utility executives lobby to delay the retirement of fossil fuel plants and bring more plants online,” to account for the power demand. “They won’t be able to find enough electricity to run all the chips,” Musk said.
How does AI fuel existing climate issues?
“Generative AI has the potential to turbocharge climate disinformation, including climate change-related deepfakes, ahead of a historic election year where climate policy will be central to the debate,” said a report by the Climate Action Against Disinformation coalition. Plus, AI, in general, is largely unregulated. “Fossil fuel companies and their paid networks have spread climate denial for decades through politicians, paid influencers and radical extremists who amplify these messages online,” the report said.
AI is also used for targeted advertising, which could encourage non-climate friendly behaviors. For example, “with fast-fashion advertising, targeted ads push a steady rotation of cheap, mass-produced clothes to consumers, who buy the outfits only to replace them as soon as a new trend arrives,” Scientific American said.
Can AI help climate change?
“Certainly, AI is at an inflection point,” Kate Brandt, Google chief sustainability officer, said in a press briefing. “Predicting the future growth of energy use and emissions from AI compute in our data centers is challenging. But if we look historically at research and also our own experience, it’s that AI compute demand has gone up much more slowly than the power needed for it.” That’s why “when we design these types of systems, we have to be mindful of the potential negative consequences and try to minimize them from the beginning,” Junhong Chen, professor of molecular engineering at the Pritzker School of Molecular Engineering and lead water strategist at Argonne National Laboratory, said to CBS News.
While data centers consume lots of energy, water-cooled data centers emit approximately 10% lower carbon emissions than air-cooled data centers, CBS News said. Companies can also find ways to make computing more efficient. A glass-half-full bonus: The technology itself can help with a variety of climate issues. For example, AI models can be used to make more efficient flight routes or help track pollution emissions. “The goal should be to avoid climate harm and instead help reduce it,” said Scientific American.