Separate from the ad hominem, it's not acceptable for someone with a science degree from MIT and a strong background in AI to take a guess at units and then confidently write that a building is using 1000x as much water as an entire city in what's presented as an authoritative work without doing a very basic sanity check first. I attribute zero malice to Hao. She was also super kind and respectful in engaging with this and I'm left with a very positive impression. I think this was an honest mistake, and have read and admired a lot of her other reporting. But this is a bigger mistake than just "guessing the wrong units" and we should be able to talk about how thousands of readers now believe a data center uses 1000x as much water as a city, and how it is actually very bad that so many people got such a wildly off impression of an important issue.

prediction, book will be changed to a statement like "in a single year, the data center could use more than THREE TIMES the amount of water consumed by the entire population of Cerrilos, a city of 88,000 people, in an entire MONTH" and everyone will keep believing what they believed before.