That's a fair point. For factual content, AI output should be verified rather than trusted blindly because hallucinations are real. I see AI more as something to challenge and verify my thinking than as something to make decisions for me.