I mean when I was pure SO debugging I would say a solid 70% of the responses did not work. Different stack different bug ect. LLMs even if they can't guess the

I mean when I was pure SO debugging I would say a solid 70% of the responses did not work. Different stack different bug ect. LLMs even if they can't guess the right answer can search the documentation really well.