This begs the question, is this wave of LLM AI anything more than a fancy mirror? They're certainly very good at agreeing with people and following along, but, as many have noted, not really useful for anything acting on their own.