5 ms·

I think their point is more on the fact that just proving something is only half the work. Intermediate results and proofs are often also of immense value. Righ

I think their point is more on the fact that just proving something is only half the work. Intermediate results and proofs are often also of immense value. Right now the output is sloppy proofs and LEAN code. This is useful, but a large part of the effort still lies ahead. And if you are a mathematician you'd rather be working on hard problems than fixing slop.

Turfie 23d ago
The main ideas for solving Navier Stokes have been exposed, it's not something buried in slop. And if it were, you could have AI write up the big ideas that actually mattered in these said slop proofs. I think they're more afraid of the current status quo. Distilling, writing textbooks, teaching, ... And currently, I could be wrong but, I see them being blind to the possibilities it brings. AI could very well be distilling maths, spreading it, and nurturing the ecosystem way better than the few elite mathematicians ever could.