Anthropic defends AI despite hallucinations May 23, 2025 By admin Although Claude once misled users, Anthropic insists AI errors are no greater than human ones and won’t block AGI.