> I don't think anybody who actually uses LLMs thinks they are general purpose AI. I would like to believe that but I have had too many conversations with peop

> I don't think anybody who actually uses LLMs thinks they are general purpose AI. I would like to believe that but I have had too many conversations with people who basically think it already is. Including in one situation of a fellow engineer. It feels like more and more "we" are in a bubble of actually having some knowledge of how this works, what the actual limitations are, and what it just is not. While there is in fact criticism of it out there, particularly around AI "art". It doesn't seem to be focused on the area we are talking about.