Comment by jesenator

I'm curious what sense you get from interacting with the best AI models (in particular Claude). From talking to them do you still chalk up their behavior to being mindless rehashing?