An AI expert finally agree with me and said one of the ways that you could determine if you were talking to an AI is you could ask it a hard question. And I thought, oh shit, that's true because AI can answer hard questions. If you said to the AI, all right, what's the capital of Albania and the AI would know but no human would. So you could immediately detect an AI just by asking any hard question. Unless they lied, you say, I know a person wouldn't have asked the answer to that. In order for AI to act like a person who would have to act like an asshole who didn't know too much then you think that's a