People usually mean that a statement or question is sufficiently ill-posed that it or any answer to it can't be right or wrong: it's tautological, or paradoxica

People usually mean that a statement or question is sufficiently ill-posed that it or any answer to it can't be right or wrong: it's tautological, or paradoxical, or otherwise doesn't clear the bar for something like https://en.wikipedia.org/wiki/Intuitionistic_logic https://en.wikipedia.org/wiki/Intuitionistic_logic. Or perhaps it relies on such a degree of hypothetical presupposition that conclusions are of no practical value outside of a bull session after work (see: Kurzweil). I don't know if that's what Pauli meant by it, but it's how it gets used today. It's especially useful in the setting of people talking about "Artificial Intelligence", because that phrase doesn't mean anything except to certain people in certain contexts. "artificial" is a pretty tricky word, "intelligence" is a really tricky word, and together they're basically impossible to define. Professional machine learning researchers talk about "performance" on "tasks". This is usually much better defined. Loosely: "Hey we got a machine to top-5 classify ImageNet pictures correctly X% more often than graduate students". That statement is right or wrong. Likewise winning at Chess or Starcraft or Go. Ditto protein folding at CASP. But "we got a machine to be intelligent/conscious"? How do you measure that? We can't even agree about what level of sophistication in an organism qualifies as "conscious", let alone "human". We can't even agree at what point an embryo becomes conscious! People really enjoy talking about AI relative with an implicit understanding that "intelligence" is "what I am", and fair play, if it's fun to talk about people can have fun talking about it. But when people start getting alarmist in an effort to generate either buzz or money and stir up a bunch of controversy around it, some of us think that "dismissive" is exactly the right attitude.