scientificamerican.com web signal

Alpöge's Claude-assisted Riemann proof clears Lean checks

TL;DR

  • Anthropic mathematician Levent Alpöge claims he used Claude to prove the Riemann hypothesis, first posed in 1859 by Bernhard Riemann.
  • MIT's Andrew Sutherland independently compiled the Lean proof and says 'the chance that anything is fishy here is approximately zero.'
  • A 'comparator run' is still pending to verify the Lean code encodes the actual theorem statement, per Imperial's Kevin Buzzard.

An Anthropic mathematician says he used the company's Claude model to prove the Riemann hypothesis, and so far independent mathematicians who have examined the work say the proof holds.

Scientific American reports that Levent Alpöge claimed the result, encoding the proof using the Lean programming language. Andrew Sutherland of MIT has separately compiled the same Lean code without issues. "The Lean proof has been compiled and independently rechecked by the Lean kernel with only the three standard axioms," Sutherland said. "Honestly the chance that anything is fishy here is approximately zero."

One check is still pending. Kevin Buzzard of Imperial College London said researchers are running a "comparator run, which will check that (a) the code does not do anything nefarious like redefine the meaning of any of the symbols in the statement." Buzzard has also laid out the reading protocol for a Lean-attested result: the first question, he said, is "what is the statement of the theorem in the Lean code, and does this correspond to the actual theorem statement."

The Riemann hypothesis, posed in 1859 by Bernhard Riemann, concerns a hidden pattern in how prime numbers are distributed. Two researchers on our roster had circulated the Scientific American piece by the time it reached us.

Shared on Bluesky by 2 AI experts