Key Points
Sam Altman claims 38,000 ChatGPT queries use as much water as one California almond.
Modern data centers consume water equivalent to large office buildings, not industrial facilities.
71% of Americans oppose data center construction due to environmental concerns.
ChatGPT has 900 million weekly users generating 2.5 billion prompts daily worldwide.
OpenAI CEO Sam Altman downplayed concerns about AI water consumption on the Sources podcast, saying 38,000 ChatGPT queries use as much water as growing one California almond. Altman acknowledged the figure came from memory and could be inaccurate. Modern data centers use about as much water as a large office building, he claimed, dismissing water concerns as a “robust meme” difficult to disprove despite public backlash.
Why the almond comparison matters
Almonds are among the most water-intensive crops grown in the U.S., requiring about 1.1 gallons of water per nut. Altman used this comparison to frame AI water use as relatively modest. He noted that people consuming 12 almonds at a time rarely feel guilty about their environmental impact, suggesting ChatGPT queries should be viewed similarly. The comparison aims to recontextualize public perception of data center water consumption.
Public opposition to data centers remains high
A Gallup report from May 2026 found 71% of Americans oppose building data centers in their area, with 70% citing environmental concerns. Half of respondents worried about resource depletion, and 18% specifically named water. A 2024 Virginia state commission report found most data centers used the same amount or less water than average large office buildings, supporting Altman’s claim.
How data centers actually use water
Data centers primarily use water for heat rejection through evaporative cooling systems. Most data centers do not reuse cooling water because evaporative systems release water as vapor into the atmosphere. After evaporation, 20% to 30% of water remains in the system as residual water containing concentrated minerals. This leftover water typically cannot be reused for cooling and is treated as wastewater or discharged per environmental regulations. A 100 MW facility consumes about 2.5 billion liters annually, roughly equivalent to 5,000 homes.
ChatGPT’s massive user base and daily queries
Despite environmental concerns, ChatGPT has 900 million weekly users worldwide generating 2.5 billion prompts daily. Monthly active users across ChatGPT and comparable AI tools like Claude reach 1.5 billion. U.S. adoption ranges from 31.3% of working-age adults to 56% of those with internet access. About one in three Americans aged 18 to 49 use large language models daily, according to a Pew Research poll from 2026. Altman acknowledged his 38,000-query figure was from memory and could be wrong, though earlier estimates suggested one ChatGPT query uses around 0.32 milliliters of water.
Final Thoughts
Altman’s almond comparison attempts to reframe AI water use as proportional to other industries, but public skepticism remains high. With 71% of Americans opposing new data centers and 900 million weekly ChatGPT users, the debate over AI’s environmental footprint will likely persist regardless of technical comparisons.
FAQs
Sam Altman estimated one query uses around 0.32 milliliters of water, though he acknowledged his 38,000-query comparison figure came from memory and could be inaccurate.
Evaporative cooling systems release most water as vapor into the atmosphere. Residual water contains concentrated minerals that scale and foul the system, making reuse impractical without treatment.
A May 2026 Gallup report found 71% of Americans oppose data center construction in their area, with 70% citing environmental concerns and 18% specifically naming water.
About one in three Americans aged 18 to 49 use large language models daily. Globally, ChatGPT has 900 million weekly users generating 2.5 billion prompts each day.
Disclaimer:
The content shared by Meyka AI PTY LTD is solely for research and informational purposes. Meyka is not a financial advisory service, and the information provided should not be considered investment or trading advice.
About Author

Danny Kontos
Co FounderDanny Kontos has been a stock investor since 2007 and co-founded Meyka in 2023. He keeps a small, focused portfolio and only moves when the numbers are hard to argue with. He has waited years on a single position before. Before Meyka, he ran a web hosting company and a mortgage lending platform, so he knows what a well-run business actually looks like under the hood. This article did not come from a news cycle. It came from someone who has been watching this space for a long time.
What brings you to Meyka?
Pick what interests you most and we will get you started.
I'm here to read news
Find more articles like this one
I'm here to research stocks
Ask Meyka Analyst about any stock
I'm here to track my Portfolio
Get daily updates and alerts (coming March 2026)