>instead of letting a few institutions wall it off for a profit.
I cannot repeat enough that that is not true, or even remotely close to true.
When ChatGPT helps me fix my toilet, saving me a $300 plumber call, I don't cut OAI a check for $300 instead.
It is unbelievably easy to harvest more than $20 of value from all the basic plans. Right now, users are overwhelmingly capturing all the value. Even if they triple the price, it's trivial for most workers to easily justify it.
We can project any future scenario we want, but right now the labs are only capturing a tiny fraction of the value they are creating. Especially for the crowd here on HN.
It's good pushback and you're right to call that out.
I think consolidation and centralized control are coming fast. Already some models are "too dangerous" to let everyone use, competition from open-weight models are being targeted as "lacking safety," distillation is under attack as "cheating," opting out of data sharing is now an enterprise-only feature, etc.
Right. But the neat things against current evolution of LLMs is that "open-weight" models are still tracking SOTA by 3-6 months, so even if all the commercial operators shut down access tomorrow, everyone would still have access to SOTA from half a year ago - which already is tremendously useful, very much in range of "saving $300 by telling you how to fix your toilet" - at the price of inference. That's because model weights are still just big bags of floating point numbers; once released, they cannot be unreleased.
Utilitarian arguments and ethics don't mix.
Sure they do. In fact, I'd tentatively claim that if you cannot easily find an utilitarian[0] argument backing an ethical one, it's a good sign the ethical argument is some flawed ideology that will be used to justify suffering and control.
--
[0] - Obviously, utilitarian argument should not be restricted to nonsense strawmen that are narrow-scoped and ignore the temporal dimension.