[thx67]

How do you know this doesn't suffer from Gell-Mann Amnesia? The first version had so many glaring errors that have been "corrected" (removed), and I don't have the energy to comb through this one.

I am highly skeptical of layperson debunking like this.

[simonw]

Andy has a good track record for writing about this. He shares plenty of credible citations - more so than most other people commenting in this space.

He also caught a major error in one of the most widely read books that helped kick off the whole data center water debate: https://blog.andymasley.com/p/empire-of-ai-is-wildly-mislead...

[thx67]

The thing that is off putting about how he uses rhetoric is that it feels like and-you deflection (tu quoque).

> Claim that a data center is using 1000x as much water as a city of 88,000 people, where it’s actually using about 0.22x as much water as the city, and only 3% of the municipal water system the city relies on. She’s off by a factor of 4500. This is the single largest error in any popular book that I’ve found on my own, and to my knowledge I’m the first person to notice it.

Yeah, she might be wrong. But a data center also using 1/5th of the water consumption of an 88k person city should still be what are debating. We also have a base rate fallacy, we don't know know how well or how poorly they are using water. Nearly all criticisms of AI data center criticisms boil (no pun intended) down to yeah, but what about almonds or rice or xyz. That isn't a healthy way to adjust how we talk about data center water consumption.

It is a classic debate tactic. Someone makes an argument, then buttresses that argument with a number. You attack the number, pretending that you didn't just correct the number, but also invalidate the original argument. We shouldn't be using these tactics to talk about the tragedy of the commons.

I worked for a hyperscaler, I poked around a bit about water usage both internally and externally and it wasn't good. There was little to no thought other than, "we can pay X to have water delivered, doesn't matter if it sourced responsibly." (to glibly paraphrase, company policy is to never write the honest part down)

Look at how hard Google fought to not have water data released in The Dalles Oregon for their DC there. Many DCs are supplied by water that is meant for humans, sourced from aquifers that took hundreds of thousands of years to fill, that are being depleted faster than they are filled, already.

I think AI is powerful tool, but we still can't give DC expansion a pass.

[simonw]

> I think AI is powerful tool, but we still can't give DC expansion a pass.

I really don't think we are. In the wider culture the idea that data centers use an absurd amount of water is baked in at this point.

It frustrates me because I think it distracts from energy usage, which is a much more real issue than water usage.

[thx67]

I'd love to see DCs scaled with renewables, they would throttle down and run slower at night. The abstraction that power is available at whatever rate you want it is an expensive myth at scale.

I defer plugging in my electric car until 10am so I know that my neighbors solar arrays are charging my car.

In Seattle, every watt saved is water saved since we are blessed with hydro.

[yellowapple]

> But a data center also using 1/5th of the water consumption of an 88k person city should still be what are debating.

Why, when that's such a minuscule amount of water in the grand scheme of things? Why focus our energy there? Why not spend that energy on the myriad many-orders-of-magnitude-worse offenders?

[dannyobrien]

So I'm not sure if it's the first version of this doc, but I ran the first version of it that's on Wayback Machine[1] with the current version to see what the differences are.

Most of the changes seem to be because Masley found one counter-example (Newton County, Georgia) where AI datacenters do seem to be increasing water costs; the only deletions AFAICS is toning down language where Masley used to say "there are no examples" to "there is one counter-example". I don't see any other major corrections that have been removed.

Here's an annotated diff of the two texts: https://bafybeie7b3zs2gqifpvn7ee7y7326wcexwnsbhnur5coymu3m6w...