I’m a big critic of OpenAI generally, I know a lot about their board members and it’s dim light between them and war criminals. With that said, I strictly appr

I’m a big critic of OpenAI generally, I know a lot about their board members and it’s dim light between them and war criminals. With that said, I strictly approve of them doing real price discovery on inference costs. Claude is dope when it doesn’t fuck up mid-response, and OpenAI is the gold standard on “you query with the token budget, you get your response”. I’ve got a lot of respect for the folks who made it stand up to that load: it’s a new thing and it’s solid AF. I still think we’d be fools to trust these people, but my criticisms are bounded above by acknowledging good shit and this is a good play.