> it's not AI that requires massive resources

no it really is. If you took away training costs, OpenAI would be profitable.

When I was at meta they were putting in something like 300k GPUs in a massive shared memory cluster just for training. I think they are planning to triple that, if not more.

Yeah for some reason AI energy use is so overreported. Using chatgpt for query does not even use two order of magnitude less energy compared to toasting a bread. And you can eat bread untoasted too if you care about energy use.

[1]: https://epoch.ai/gradient-updates/how-much-energy-does-chatg...

On the contrary, there's absolutely a reasoned, principled position here. Pike isn't a hypocrite for creating a Markov chain bot trained on the contents of an ancient public domain work and the contents of a single usenet group, and still complaining about modern LLMs; there's a huge difference in legality and scale. Modern LLMs use orders of magnitude more resources and are trained on protected material.

Now, I don't think he was writing a persuasive piece about this here, I think he was just venting. But I also feel like he has a reason to vent. I get upset about this stuff too, I just don't get emails implying that I helped bring about the whole situation.