News for hackers. Stuff that matters.
Re:Anthropic employee claims general proof of the Riemann Hy... by imhoguy on Friday November 20, 2026 @10:09AM (#50575729)
CS layman folk here. How one is going to prove 250K page argument? Just single sign of hallucination can make it false, no?
Re:Anthropic employee claims general proof of the Riemann Hy... by tasuki on Friday November 20, 2026 @10:19AM (#50575901)
Re:Anthropic employee claims general proof of the Riemann Hy... by NoDodgeQuestion on Friday November 20, 2026 @10:20AM (#50575933)
AI proof in lean, can verify by computer. Mistake can only be compiler bug or state problem wrong.
Re:Anthropic employee claims general proof of the Riemann Hy... by well_ackshually on Friday November 20, 2026 @10:45AM (#50576448)
Mathematicians, famously known for being extremely precise in their specification of problems and absolute lack of errors writing down proofs.

It's always, always, always stating problem wrong.

Re:Anthropic employee claims general proof of the Riemann Hy... by dgellow on Friday November 20, 2026 @11:29AM (#50577280)
No, it can also be that the proof doesn’t actually prove the thing you think it does, and a lot of other variant of that issue. Lean doesn’t make it magically valid
Re:Anthropic employee claims general proof of the Riemann Hy... by Epa095 on Friday November 20, 2026 @10:22AM (#50575980)
Its formalized in lean, which verifies each step. The steps themselves are relatively easy, and can be mechanically verified (think "A and A->B thus by modus ponens B").

Traditionally writing verified proofs is a hard undertaking, exactly because you need to write out a lot of steps "normal human" mathematics jump over, but with AI its easier.

Re:Anthropic employee claims general proof of the Riemann Hy... by vintermann on Friday November 20, 2026 @10:29AM (#50576140)
At what point do we start worrying that it's found a bug in Lean instead?
Re:Anthropic employee claims general proof of the Riemann Hy... by suddenlybananas on Friday November 20, 2026 @10:49AM (#50576517)
In principle you can run Lean with different kernels and so there would have to be a bug in all of them for the proof to be faulty I believe.
Re:Anthropic employee claims general proof of the Riemann Hy... by cma on Friday November 20, 2026 @11:02AM (#50576776)
Still possible, but Lean ships with formally verified kernels as of a few weeks ago.
Re:Anthropic employee claims general proof of the Riemann Hy... by thejokeisonme on Friday November 20, 2026 @11:03AM (#50576794)
Right now.
Re:Anthropic employee claims general proof of the Riemann Hy... by d-u-d-e on Friday November 20, 2026 @11:05AM (#50576835)
The Lean kernel should be easy to verify by hand. We can ask the same question with humans though. What if a result that was used to prove some other result turns out to be wrong just because whoever reviewed it neglected a small detail?
Re:Anthropic employee claims general proof of the Riemann Hy... by dgellow on Friday November 20, 2026 @11:28AM (#50577251)
That already happened multiple times, it’s a known worry
Re:Anthropic employee claims general proof of the Riemann Hy... by est on Friday November 20, 2026 @10:36AM (#50576268)
> The steps themselves are relatively easy, and can be mechanically verified

They aren't

Re:Anthropic employee claims general proof of the Riemann Hy... by cubefox on Friday November 20, 2026 @10:24AM (#50576022)
Small errors in proofs can often be fixed easily. So it's not that the whole thing would be useless. Also, he has a Lean proof, which rules out these small errors.
Re:Anthropic employee claims general proof of the Riemann Hy... by adgjlsfhk1 on Friday November 20, 2026 @10:27AM (#50576098)
Step 1 will be making a much shorter argument.
Re:Anthropic employee claims general proof of the Riemann Hy... by criddell on Friday November 20, 2026 @11:01AM (#50576762)
[delayed]