This to me is what I see want to see as the fallacy in the anti-ChatGPT argument, but at the same time have to acknowledge as the real danger. I think AI paired

This to me is what I see want to see as the fallacy in the anti-ChatGPT argument, but at the same time have to acknowledge as the real danger. I think AI paired with a human, as you are describing, can be more effective and efficient than either on their own. However, I've never seen it stay that way in commercial applications. So I'm afraid that the smallest drive towards cost-cutting will lead to removal of the human in the loop or at least tighten throughout expectations so much that the human intervention is in practice more often missing than not. Not sure what to do about this. I want ChatGPT as a tool available to me, but I don't want any service provider I interact with to use it because I know they'll fuck it up eventually and I'll get worse service.