Fine. I’m happy to assume that, in my hypothetical, we observe that it’s always very nice and hard to make not-nice. I claim that a bunch of people would still skeptically ask “but how is this relevant to future models?”
Fine. I’m happy to assume that, in my hypothetical, we observe that it’s always very nice and hard to make not-nice. I claim that a bunch of people would still skeptically ask “but how is this relevant to future models?”