That’s the whole point, though. The need to make things interpretable from the center is what it means to “see like a state.” A society doesn’t need to be interpretable to exist. Large groups of people can relate to one another peerwise.
HN user
rosecross
My first thought was this could be great for alignment, e.g., M_aligned = M - M_toxic.
Turns out, that’s exactly what they’ve explored in the paper.
Great stuff.
Interests don’t stay in boxes. They grow tentacles that reach up and down the chain of production. They bump up against adjacent fields and they make the whole world look a little different. A single curiosity will evolve and change form over time. The beginning of it might not be boring to others, but if you take it far enough the end probably will be.
He mentions that for certain projects you’ll need to be good at managing others, or you shouldn’t attempt them at all.
Guess you didn’t read to the end
Everyone likes movies, but that doesn’t mean they want to discuss cinematography all night.