“Rotten worldview:” Altman sparks outrage by comparing training AI to raising a child
OpenAI CEO Sam Altman has tried to justify the extensive energy use of artificial intelligence (AI), comparing it to the resources needed for human development. His remarks didn’t go down well with many commenters online.
Altman responded to criticism about AI’s energy usage by saying it’s unfair to compare the energy it takes to train an AI model relative to how much it costs humans to do one inference query.
Speaking at an event hosted by the Indian Express, OpenAI CEO compared training AI systems to raising a child to adulthood, noting that training a human is also energy-intensive.
“It takes like 20 years of life and all of the food you eat during that time before you get smart,” he said.
Altman went so far as to compare training models to the evolution of a hundred billion people who have ever lived on Earth and learnt “not to get eaten by predators and figure out science.”
“The fair comparison is: if you ask ChatGPT a question, how much energy does it take once its models are trained to answer this question versus a human? Probably AI has already caught up on an energy efficiency basis if we measure that way,” Altman said.
Technology experts and spiritual leaders condemned the parallels between the energy consumed by AI data centers and the resources required to raise a human.
Pastor David Fairchild called it anthropology, “where humans are basically inefficient meat computers that you have to pour food and years into before they become useful.”
“If your worldview can look at a child growing into an adult and describe it as energy spent to train intelligence, you haven’t said something profound. You've revealed a horrifically rotten worldview,” Fairchild wrote on X.
Ewan Morrison, an author and cultural critic, said that Altman has become “an article in the Onion,” a satirical media outlet.
“Data centers use a lot of energy – yes, but so do humans. Therefore, we must feed data centers proteins, carbs, and greens as well as millions of tons of fresh water,” Morrison joked in his X post.
According to Laura Kittel, PhD, a human rights advocate, any technological advancement that portrays people “as an inefficiency to eliminate rather than serve” misses the point of its existence and should be seriously re-evaluated.
Data center electricity use is set to double
AI companies have been increasingly criticized for their heavy energy consumption. Data centers require vast amounts of water, mostly potable, for cooling systems to prevent servers from overheating.
Altman has previously said that ChatGPT uses less than 1/15 teaspoon for an average query.
However, independent studies paint a slightly different picture. According to the Environmental and Energy Study Institute, large data centers can consume up to five million gallons per day, equivalent to the water use of a town populated by up to 50,000 people.
The conversation on this topic is live. Join in the discussion.
The Institute’s report notes that with larger and new AI-focused data centers, “water consumption is increasing alongside energy usage and carbon emissions.”
The International Energy Agency (IEA) report states that a typical AI-focused data center consumes as much electricity as 100,000 households. However, the largest data centers that are currently under construction are estimated to consume 20 times as much.
By 2030, global data center electricity consumption is set to more than double, reaching 945 TWh, and slightly exceed Japan’s total electricity consumption today.
Unlock more exclusive Cybernews content on YouTube.
