AIThis post was created with the assistance of artificial intelligence (AI).

🔍 Read the full analysis: How Was Claude’s Help In Riemann Hypothesis Research Checked? on ThorstenMeyerAI.com

Before you orderOffer from Amazon

Get the latest gadgets delivered free with Prime

  • Fast, free delivery on millions of items
  • Prime Video, Amazon Music and more included
  • Member-only deals all year
Start your free Prime trial Free trial for eligible customers · Cancel anytime
As an affiliate, we earn on qualifying purchases.

TL;DR

Levent Alpöge, an Anthropic mathematician, said on November 20 that he believes he and Claude proved the Riemann hypothesis and the stronger Generalized Riemann Hypothesis. The claim is supported by a large Lean proof archive, but expert review is at an early stage: an independent compile was reported, while a comparator test and human understanding remain outstanding.

Mathematician Levent Alpöge, who works at Anthropic, posted a note on November 20 saying he believes he and Claude have proved the Riemann hypothesis, a major unsolved problem in mathematics. The claim is not an accepted result: the proof has not appeared in a refereed journal or received broad mathematical review, and specialists were still checking the formal code on November 21.

Alpöge’s four-page note, posted on his personal website, also claims that the approach extends to the Generalized Riemann Hypothesis. The note itself does not present a conventional, readable proof. Instead, it links to a formal proof archive in Lean, a programming language and proof assistant that can check whether mathematical steps follow from stated assumptions and rules.

Computer scientist Scott Aaronson put the archive at 17.4 million lines of Lean code. Alpöge said a human-readable version would exceed 250,000 pages and that he chose to publish the Lean files instead. The source account says MIT mathematician Andrew Sutherland told Scientific American the code compiled independently “with only the three standard axioms.” That is an encouraging technical check, but it does not settle whether the code formalizes the intended theorem. Kevin Buzzard said a “comparator run” was still awaited—a further test intended to check that the formalized result matches the real mathematical statement.

The division of work is described in Alpöge’s account, not independently established. He says he set the strategy and that Claude handled substantial technical work, writing that “Claude far outpaced me near the finish line.” He also says he had “little mastery” of one of the areas involved. The note does not identify which Claude model was used. Anthropic had not announced the claim on its news page as of November 21, and Nature reported that the company did not immediately respond to a request for comment.

At a glance
reportWhen: Claim posted November 20, 2026; initial…
The developmentLevent Alpöge posted a claim that he and Claude proved the Riemann hypothesis, prompting initial checks of a 17.4-million-line Lean archive.

Explainer · Riemann hypothesis claim

From claim to accepted proof: six checks, and where this one stands

What the hypothesis says, and the path every proof claim has to take before mathematicians accept it.

The claim in one picture

0 1/2 1 real part

Riemann’s guess (1859): every non-obvious zero of the zeta function lies on this one line. If true, the primes are spread out as evenly as possible.

The checks a proof claim must pass

  1. 1Claim postedA 4-page note plus Lean code, November 20, 2026✓ done
  2. 2Code compiles elsewhereIndependent compile reported by an MIT mathematician✓ reported
  3. 3Comparator runRight statement? Only standard axioms? Several Lean kernels?… pendingWe are here · November 21, 2026
  4. 4Humans understand itA readable proof that experts can follow○ open
  5. 5Journal publicationA refereed mathematics journal of worldwide repute○ open
  6. 6Two years + general acceptanceWhat the Clay prize rules require before a $1 million award○ open

Sources: Clay Mathematics Institute (Riemann Hypothesis page; Millennium Prize rules); Levent Alpöge, remark.pdf (November 20, 2026); Scientific American, November 20, 2026 (independent compile; comparator run awaited); Kevin Buzzard, Xena Project, November 12, 2026 (the checks). Status as reported in these sources by November 21, 2026.

Graphic: Powered by Thorsten Meyer AI

Why a Formal Proof Could Matter

A valid proof would resolve a question that has shaped number theory since Bernhard Riemann posed it in 1859. The hypothesis concerns the locations of certain zeros of the Riemann zeta function. If true, it implies strong regularity in how prime numbers are distributed and supports many results that are currently conditional on the hypothesis.

The claim also puts AI-assisted mathematics under an unusually public test. A proof assistant can check formal steps at a scale no person could readily inspect line by line. But a successful software check is not the same as mathematicians understanding the argument, verifying that the formalization captures the intended theorem, and accepting the result after scrutiny. Those are separate questions, and the reports available so far do not resolve them.

The Clay Mathematics Institute lists the Riemann hypothesis as one of its Millennium Prize Problems, with a $1 million prize. A posted claim does not start the prize process: the institute’s rules require publication in a refereed journal, at least two years of scrutiny, and general acceptance by the mathematics community.

Products Worth Considering

As an affiliate, we earn on qualifying purchases.

From Riemann’s Question to Lean

The Riemann hypothesis says that all the “non-obvious” zeros of the zeta function have real part one-half. The conjecture is closely connected to how regularly prime numbers appear among the whole numbers. It has been checked computationally for the first 10 trillion solutions, according to the Clay Mathematics Institute, but checking any finite number of cases cannot prove a statement about infinitely many cases.

Lean offers a way to encode a proof in a precise formal language and have software verify its logical steps. In this case, the archive’s size makes that facility useful, but it also creates a practical challenge: reviewing whether the code proves the stated theorem and whether the result can be understood and assessed by other mathematicians. Terence Tao described the submission as an “(enormous) Lean-only proof,” while the source account says Andrew Granville called the released note “garbled.”

The public evidence is still limited to the note, the linked code, and early comments from mathematicians. Alpöge’s description of Claude’s contribution is important evidence about how the work was produced, but it remains the author’s account rather than an independently documented breakdown of each contribution.

“I believe I have, working with Claude, discovered a proof of the Riemann Hypothesis.”

— Levent Alpöge

What the Early Checks Do Not Establish

The reported independent compilation is a preliminary technical result, not a mathematical endorsement. The comparator run was still pending in the source account, and the code’s full review had not been completed. It is not yet clear whether independent specialists will confirm that the formal proof proves the Riemann hypothesis as stated, or whether the claimed extension to the Generalized Riemann Hypothesis also holds.

There is also no public, accessible exposition of the argument in the material described. Alpöge says a document exceeding 250,000 pages exists, but he did not post it. The model version, the precise sequence of Claude’s contributions, and how much of the proof can be independently understood by mathematicians remain unknown. No journal publication or formal acceptance had been reported by November 21.

The Checks That Could Change the Claim

The immediate next step is the reported comparator run, followed by further inspection of the Lean archive by people who can assess both its formal validity and its connection to the stated theorem. Any defects, corrections, or successful independent reproductions would materially change how the claim is evaluated.

Beyond code checks, the proof would need to enter the usual process of mathematical scrutiny. Under the Clay Institute’s rules, a prize claim requires refereed publication, at least two years of scrutiny, and general acceptance. Until those stages occur, the accurate description is that Alpöge has posted a substantial formal proof claim involving Claude—not that the Riemann hypothesis has been settled.

Key Questions

Has Claude proved the Riemann hypothesis?

Not as an accepted mathematical result. Alpöge says he believes he and Claude found a proof, but independent review is incomplete and no journal or mathematical body has accepted it.

How is the proof being checked?

The argument is encoded in Lean, which checks formal logical steps. An independent compile was reported, while a comparator run and broader expert scrutiny were still pending.

What did Claude contribute?

According to Alpöge, he guided the strategy and Claude carried out substantial technical work. That division comes from Alpöge’s account; the note does not identify the Claude model used.

What would the claim mean if verified?

A proof would resolve a central question about the zeta function and the distribution of prime numbers. It could also establish results in number theory that currently depend on the hypothesis.

Does the claim qualify for the Clay prize now?

No. The Clay Mathematics Institute requires refereed publication, at least two years of scrutiny, and general acceptance before a prize claim meets its stated conditions.

Source: ThorstenMeyerAI.com

You May Also Like

A Deep Dive Into Valve’s Gaming Hardware Strategy And The Missing Steam Machine

An analysis of Valve’s hardware plans, the absence of a dedicated Steam Machine, and what it reveals about their strategy.

From One Prompt to Nine Games: What an AI Did With “Make Your Own Stickman Game”

It started with one loose prompt and a rhythm stick-fighter. One day later there were nine games, seven venues, a VERSUS mode and an animator, all free in the browser and all made of code.

Claude Sheets: A User’s Guide To Features, Whiteboards, And Pricing

Anthropic launched Claude Sheets and Whiteboards in beta on Nov. 10. Here are the supported plans, pricing, features and current limits.

Harvard’s Claude Access Expansion: Student Uses In Class And Beyond

A Harvard Crimson headline reports students are using expanded Claude access in classes and personal projects; key program and impact details remain unavailable.