Of course you are wrong. "IEA's models project that data centres will use 945 terawatt-hours (TWh) in 2030, roughly equivalent to the current annual electricity consumption of Japan." [1] And don't forget the recent announcement by Fermi America to build a 6 gigawatt nuclear datacenter in Texas. And what are we getting in return:
(1) AI is primarily used, by and far, to accelerate consumerism.
(2) AI has very few applications that are actually solving the world's problems (the world's problems are mostly nontechnical problems). The cited ones of scientific or medical are either too abstract, or else they are likely to make the problem worse. And in the case of medical, maybe the AI applications will help a few hundreds of thousands of people – but hurt/kill many more with its contribution to climate change. So not worth it.
I consider those who promote AI to be enemies of humanity and biological life since they are using a commodity that we should be using less of with reckless abandon.
Nuclear is good though, it's the single densest source of energy in the world (and, well, outside of it too).
I don't get your argument though, where has it "accelerated consumerism?" And you can't be serious about its medical applications, I doubt more people would die from its incremental climate change effects (as compared to, say, car accidents, much less car pollution in cities) as compared to being saved from death. Even outside of the scientific fields, AI sure is solving a lot of my problems, saying one is an "enemy of humanity" is the sort of hyperbole I'd only see on HN.
> Nuclear is good though, it's the single densest source of energy in the world (and, well, outside of it too).
It would be good if the energy were used in critical applications. Wasted energy is stil wasted.
> I don't get your argument though, where has it "accelerated consumerism?"
Are you serious? It is making people richer, allowing people to make products faster, including software. Of course that makes consumption faster.
> And you can't be serious about its medical applications, I doubt more people would die from its incremental climate change effects (as compared to, say, car accidents, much less car pollution in cities) as compared to being saved from death.
No, I am serious. And it will get much worse in the future. Check out [1], [2], etc.. And lets not forget the increased storms, flooding, etc. We are all responsible for that. But those that use electricity like big tech should be held especially responsible because they encourage the bad behaviour and use resources directly.
> Even outside of the scientific fields, AI sure is solving a lot of my problems, saying one is an "enemy of humanity" is the sort of hyperbole I'd only see on HN.
Solving your problems doesn't really mean solving the world's problems, just like making the top 10% richer doesn't make the world a better place. I absolutely consider them the enemy of humanity.
(1) AI is primarily used, by and far, to accelerate consumerism.
(2) AI has very few applications that are actually solving the world's problems (the world's problems are mostly nontechnical problems). The cited ones of scientific or medical are either too abstract, or else they are likely to make the problem worse. And in the case of medical, maybe the AI applications will help a few hundreds of thousands of people – but hurt/kill many more with its contribution to climate change. So not worth it.
I consider those who promote AI to be enemies of humanity and biological life since they are using a commodity that we should be using less of with reckless abandon.
1. https://hardware.slashdot.org/story/25/04/10/2019233/data-ce...