Stupid question: Why can't models be trained in such a way to rate the authoritativeness of inputs? As a human, I contain a lot of bad information, but I'm awar

Stupid question: Why can't models be trained in such a way to rate the authoritativeness of inputs? As a human, I contain a lot of bad information, but I'm aware of the source. I trust my physics textbook over something my nephew thinks.