Generative AI’s environmental costs are “closely guarded corporate secrets,” scientist criticizes



summary
Summary

AI has high environmental costs that are largely unknown, says Kate Crawford, a researcher at USC Annenberg and Microsoft Research who focuses on the social implications of artificial intelligence.

In addition to energy, generative AI systems require large amounts of fresh water to cool their processors and generate electricity, Crawford says.

She cites estimates that by 2027, global water use for AI could be equivalent to half of the United Kingdom’s consumption.

In West Des Moines, Iowa, a data center for OpenAI’s GPT-4 accounted for about 6 percent of the county’s water use in July 2022. According to their environmental reports, Google and Microsoft’s water use increased by 20 percent and 34 percent, respectively, within a year.

Ad

Ad

Instead of relying on new technologies like nuclear fusion, as recently proposed by OpenAI CEO Sam Altman, Crawford calls for pragmatic measures to limit AI’s environmental impact. These include prioritizing energy efficiency, building more efficient models, and redesigning data centers.

“Rather than pipe-dream technologies, we need pragmatic actions to limit AI’s ecological impacts now,” says Crawford.

Environmental costs of generative AI are “closely guarded corporate secrets”

The full environmental costs of AI are “closely guarded corporate secrets,” according to Crawford. Current figures are based on research, limited company reports, and data published by local governments. They paint an incomplete picture, and “there’s little incentive for companies to change.”

Crawford hopes that US Democrats, led by Senator Ed Markey of Massachusetts, can push through the Artificial Intelligence Environmental Impacts Act of 2024, which was introduced on February 1.

The bill would require the National Institute of Standards and Technology to work with scientists, industry, and civil society to develop standards for assessing the environmental impacts of AI and create a voluntary reporting framework for AI developers and operators.

Recommendation

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top