Glen’s take · Why we chose this story
Primary source brookings.edu
Open the original on brookings.eduIn short the story, from brookings.edu
Brookings policy brief: AI data centers projected to consume 68 billion gallons of water annually by 2028, up from 17 billion in 2023. Large facilities use 5 million gallons per day — equivalent to a town of 50,000 people. Source · brookings.edu
The read Glen’s take, in his words
I want to share numbers that stopped me. Brookings just documented how much water AI data centers are projected to consume — and the scale is staggering. Seventeen billion gallons a year in 2023. Sixty-eight billion projected by 2028. Source · brookings.edu
A single large facility now uses five million gallons a day. That's the equivalent of a town of fifty thousand people — on one data center.
What this Brookings analysis captures is that it's not theoretical. It's the infrastructure curve we're already on. Every new AI model, every new cloud workload, adds to the draw — and almost none of the communities hosting these facilities understood what they were signing up for.
I wanted you to see this. The math is the story.
Questions answered from this page
What is this story about?
Brookings policy brief: AI data centers projected to consume 68 billion gallons of water annually by 2028, up from 17 billion in 2023. Large facilities use 5 million gallons per day — equivalent to a town of 50,000 people.
Why did Glen Kerby choose this story?
We need to share the impact that the Ai universe is having on the world of water.
Where can I read the original?
It was published by brookings.edu as “AI, data centers, and water”, at https://www.brookings.edu/articles/ai-data-centers-and-water/.
Go to the source their work, in full
“AI, data centers, and water”
Published by brookings.edu. Read it where it was reported.
