Back to articles

Anthropic Mathematician Says Claude Helped Prove the Riemann Hypothesis

3 min read

Anthropic mathematician Levent Alpöge says he has found a proof of the Riemann hypothesis with help from Claude. The problem, which concerns the distribution of prime numbers, is among mathematics’ best-known open questions and one of the seven Millennium Prize Problems. If the proof holds up, it would be a major development for mathematics and AI-assisted research. For now, however, the careful description is that a proof has been proposed—not that the problem has been solved.

What is known so far

  • Alpöge posted the proposed proof on his personal website and says Claude played an integral role in developing the argument.
  • He says the reasoning has been formalized in Lean and suggests that the method may extend to the Generalized Riemann Hypothesis and potentially further.
  • The work has not been independently verified or formally accepted by the Clay Mathematics Institute or the mathematical community. Anthropic has not confirmed its involvement; at the time of The Verge’s report, the company had not responded to its inquiry.
  • Under the institute’s rules, a solution must be published and withstand at least two years of scrutiny, while receiving general acceptance among mathematicians, before it can be considered for the prize.

The central question is not simply whether Claude was involved, but whether other mathematicians can check every step and confirm that the argument is sound. Proof assistants such as Lean can help express mathematical reasoning in a machine-checkable form. That is valuable, but formalization alone does not amount to broad mathematical acceptance. Researchers still need to examine the definitions, assumptions, individual steps, and whether the formalized version faithfully represents the original argument. The report does not include enough detail for readers to assess the proof’s correctness.

The announcement comes amid heightened attention to AI’s role in mathematics. The Verge also describes OpenAI’s earlier claim of progress on the Navier-Stokes problem, a result that has yet to receive formal recognition and has faced controversy over research practices. Alpöge had previously said he used Claude to solve another conjecture. Such announcements suggest that AI tools are being applied to difficult mathematical questions, while also putting disclosure, research credit, and verification under greater scrutiny.

Verification will determine the significance

If independent experts eventually confirm the proof, it could have far-reaching consequences and become an important case study in AI-assisted mathematics. Until then, a researcher’s claim should not be treated as an established theorem, nor as evidence that Claude independently solved the problem. Evaluating AI’s scientific capabilities requires distinguishing model assistance, formal verification, and results accepted by experts. The next steps to watch are whether the full argument can be scrutinized publicly, whether independent teams can reproduce its verification, and whether it gains broad acceptance among mathematicians.

Source: The Verge AI

Comments

Checking sign-in status...

Loading comments...

Related articles