Google, supposedly full of AI experts, doesn't understand what current AI is good for and its limitations. A lot of AI critics (and enthusiasts) expect AI to be

Google, supposedly full of AI experts, doesn't understand what current AI is good for and its limitations. A lot of AI critics (and enthusiasts) expect AI to be a know-it-all that never fails and provides authoritative factual answers. But current technology AI isn't good at all for this. I use AI everyday for giving me opinions, starting points for re/search, reviewing my work and brainstorming. AI is fabulous for that and was a life changer for me. Hallucinations are actually welcome because they provide new perspectives. And when AI fails me, that's not different from getting an unhelpful perspective from a friend. I wish more people were focusing on what LLMs do best: augmenting human abilities rather than replacing them.