Company worried higher numbers could damage its reputation
Amazon strategised about ways to keep the public in the dark over the true extent of its data centres’ water use, a leaked internal document reveals.
The biggest owner of data centres in the world, Amazon dwarfs competitors Microsoft and Google and is planning a huge increase in capacity as part of a push into artificial intelligence. The Seattle-based company operates hundreds of facilities worldwide, with many more planned despite concerns over how much water is being used to cool them.
Amazon’s data centres were projected to use 7.7 billion gallons of water a year by 2030, according to the leaked strategy memo, which was circulated within the company in 2022. The $2.4 trillion corporation defends its approach to water usage and has taken steps to improve water efficiency.
But while Microsoft and Google regularly publish their water consumption, Amazon has never publicly disclosed how much water its giant server farms consume. In the leaked document, Amazon executives warned that transparency was “a one-way door” and advised keeping its projections confidential, even as they feared inviting accusations of a cover-up.
“Amazon hides its water consumption,” was one hypothetical headline the authors warned of.
Amazon data centres in Arizona, where their water use has been controversial (picture: Tedder)
When designing a campaign for water efficiency, Amazon Web Services (AWS), the company’s cloud computing division that oversees its data centres, noted that it would be harder to reach its internal target if its calculations included “secondary” use—water used in generating the electricity to power its data centres, according to the document.
Instead, Amazon officials opted to use only the relatively smaller figure of primary use, 7.7 billion gallons, when calculating progress towards its target because of “reputational risk”, fearing bad publicity if the full scale of Amazon’s consumption was revealed, the document shows.
“In environmental science, it is standard practice to include both to more accurately capture the true water cost of data centres,” said Shaolei Ren, associate professor of electrical and computer engineering at the University of California, Riverside.
... continue reading