I don't think he is. That's sort of his point.

I like this guy’s backstory[0].

I grew up in Africa. The poverty I saw, as a child, was foundational in my own personal development.

There are some places in the US, that have that kind of poverty, but I have not seen them, with my own eyes.

I have family that dedicated most of their life to fighting poverty (with very limited success). They believe that poverty is probably the single biggest problem in the world, today. Almost every major issue we face, can be traced back to poverty.

Income inequality is one thing, but hardcore poverty, as described by the author, is a different beast, and creates a level of desperation that is incredibly dangerous.

[0] https://blog.ctms.me/about/

Tip for AI skeptics: skip the data center water usage argument. At this point I think it harms your credibility - numbers like "millions of liters of water annually" (from the linked article) sound scary when presented without context, but if you compare data centers to farmland or even golf courses they're minuscule.

Other energy usage figures, air pollution, gas turbines, CO2 emissions etc are fine - but if you complain about water usage I think it risks discrediting the rest of your argument.

(Aside from that I agree with most of this piece, the "AGI" thing is a huge distraction.)

UPDATE an hour after posting this: I may be making an ass of myself here in that I've been arguing in this thread about comparisons between data center usage and agricultural usage of water, but that comparison doesn't hold as data centers often use potable drinking water that wouldn't be used in agriculture or for many other industrial purposes.

I still think the way these numbers are usually presented - as scary large "gallons of water" figures with no additional context to help people understand what that means - is an anti-pattern.

Awesome that there are releasing this paper and the associated data. I hope they'll do this regularly, so that changes can be tracked.

One thing I hope they'll correct going forward is inclusion of API usage. Anecdotally, I only use Anthropic models via Cursor. So none of that usage shows up in here. I'd expect that specialized tools/interfaces like Cursor will grow and thus more usage will shift to API. It would be a shame to miss out on that in the data set.

I tried doing some back of the envelope math but it doesn’t come out in favor of drones.

The CL-415 that you shared in the link has a reported capacity around 6100 kg of water. When I searched for “heavy lift drones” I got the Draganfly Heavy Lift, which can reportedly carry 30 kg but only for 18 minutes of flight time (according to the company’s web site). I think the plane’s flight time is around 3 hours but obviously the math is a little more complicated than that.