Sam Altman Defends Use of AI Power Against Humans, Spark Debate

Sam Altman goes back growing criticism over environmental tax of AI The OpenAI The manager dismissed the claims about AIs water used as a “fake” comparison drawn between the electricity required to run AI systems and the energy required to develop human intelligence.

Statistics suggest that the tools are similar ChatGPT eat many liters water each question is “absolutely crazy” and “out of touch with reality,” Altman said in a February 20 interview with The Indian Express on the sidelines of the AI ​​Impact Summit in New Delhi. Last year, Altman said ChatGPT uses 0.000085 liters of water per query—about one-fifteenth of a teaspoon—though he didn’t explain how he calculates that number.

The AI water footprint stems largely from the need for evaporative cooling systems used to keep data center hardware from overheating. But Altman argued that companies like OpenAI no longer directly manage such cooling processes. Many AI developers, he said, are turning to cooling systems that recirculate fluids rather than continuing to pump new materials. Meanwhile, tech giants like Microsoft, Meta, Google and Amazon have promised to fill in more. water rather than withdrawing in 2030.

Nevertheless, data centers continue to drink water at a fast pace. Total related to AI water cooling consumption will reach 23.7 cubic kilometers by 2025, a 38 percent increase over 2020, and is expected to more than triple over the next 25 years, according to a January report from Xylem. Despite the industry’s stance on alternatives, the report found that 56 percent of data center capacity still relies on some form of evaporative cooling.

Altman was very measured when it came to the use of electricity. “What is right, however, is the use of force,” he said. “We need to move to nuclear, wind and solar as quickly as possible.”

Last April, the International Energy Agency reported that data centers will account for approximately 1.5 percent of global electricity consumption by 2024. Their energy consumption is increasing at a rate four times faster than total electricity demand and is expected to more than double by 2030.

As a result, large technology companies are pursuing data center agreements tied to other energy sources, including nuclear power, to ease pressure on the grids. Altman, who once led Y Producerhe personally invested himself in nuclear activities such as That’s rightbuilding small nuclear plants, and Helionaimed at commercializing nuclear fusion.

The OpenAI CEO also pointed out that critics underestimate the power required to develop human intelligence. “People talk about how much energy it takes to train an AI model compared to how much it costs a human to do one query,” he said. But it also takes a lot of energy to train someone—it takes 20 years of life and all the food you eat in that time before you start.”

A more appropriate comparison, he suggested, would be to measure the power used by a fully trained AI model to answer a question against that used by a human doing the same task. “Perhaps AI is already playing a role in the efficiency of energy measured in that way.”

The comments quickly sparked a debate online about whether such comparisons are appropriate. “You say it’s a really big spreadsheet and the child is morally equivalent,” wrote Matt Stoller, research director of the American Economic Liberties Project, in a post on X. Sridhar Vembu, founder and chief scientist of software company Zoho Corporation, also disputed the OpenAI executive’s statements. AI should “quietly take a back seat” instead of dominating our lives, billionaire tells X. “I don’t want to see a world where we equate a piece of technology with a person.”