News

Anthropic mathematician says Claude helped prove Riemann hypothesis

Anthropic mathematician Levent Alpöge says work with Claude produced a proof of the Riemann hypothesis, but independent examination of the Lean archive has not begun.

D
Nov 20, 2026 · 1 min read

Anthropic mathematician Levent Alpöge says he discovered a proof of the Riemann hypothesis while working with Claude, but the claim has not been independently established. The result would settle a 167-year-old problem that the Clay Mathematics Institute still lists as unsolved.

Alpöge released a four-page announcement and a Lean archive, describing the result as a belief rather than a verified resolution. He says he and Claude formalized the argument in Lean without assumptions. The announcement links to 14 downloadable archive parts, but it does not include a conventional prose proof. Alpöge said he intended to understand, simplify and write the argument properly. The announcement does not identify the Claude model or version used.

Lean is a proof assistant used to encode arguments for machine checking. Other AI-assisted mathematics formalization projects provide context for that approach, but do not establish whether Alpöge’s archive is correct.

Clay describes the Riemann hypothesis as the assertion that every non-obvious zero of the Riemann zeta function has real part one-half. Alpöge also claims his method resolves the generalized Riemann hypothesis and could extend to the extended Riemann hypothesis with more work. Those assertions remain unverified alongside the central claim.

The validation work remains substantial. State of Proof recorded the submission as an intake and said its independent examination had not begun. Reviewers would need to confirm that the Lean declarations correspond to the standard Riemann hypothesis and Dirichlet generalized Riemann hypothesis, trace the dependency tree, and determine whether imported material introduces assumptions that change the claim.

Even a technically valid archive would not immediately qualify as a Millennium Prize solution. Clay’s rules require publication in a qualifying outlet, a wait of at least two years and general acceptance in the global mathematics community before the institute will consider a proposed solution.

More news