@searls This. I’m seeing the same thing in any domain where there is expertise—the experts see the problems in LLM generated content (relevant to that domain); everyone else thinks it works just fine—they don’t see the gaps & errors. 🙄