Computer people: I keep wondering whether LLMs are especially vulnerable to hallucinations (or whether I am especially aware of them, because I spend my life writing), or whether this is an AI vulnerability across applications - coding, science research, wherever else it is used?
2 days ago