- It is also unclear if this proof is something a human couldn't have come up with
I don’t disagree, but this is not established science.
We actually don’t know if it’s easier to verify a proof than to come with a proof.
It’s one of the fun things that lead to P!=NP if it’s correct.
Again, probably correct but if you could prove it that would be groundbreaking.
That's clearly not true here.
If it was that easy, OpenAI would have already announced the result weeks ago.
So if we stick a 10 year old in front of Claude and instruct them to type in “create a proof of the Riemann hypothesis”, as long as they keep pressing return, eventually a proof will pop out?