> A human proof won't just be logically correct, it will be cognitively distilled and coherent. It may still take years to understand it, but the logical flow will be straightforward, not obfuscated.
https://en.wikipedia.org/wiki/Inter-universal_Teichmüller_th... seems like a counterpoint, but IANAM. (I am likely cherrypicking the far end of the bell curve re: straightforward here)
Isn't this controversial, to say the least?
I wait for AI to say something about this :)
Someone at OpenAI, please, work on this.
Not a counterpoint, actually case in point, because:
>Mochizuki and a few other mathematicians claim that the theory indeed yields such a proof but this has so far not been accepted by the mathematical community.
Proof can't be understood, proof doesn't matter.