> Except a "good product" is safe. Depends on how you define "safe". The kind of "safe" we get from OpenAI today seems to be mostly censorship, I don't think w

> Except a "good product" is safe. Depends on how you define "safe". The kind of "safe" we get from OpenAI today seems to be mostly censorship, I don't think we need more of that.