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

🔍 Read the full analysis: An Anthropic Mathematician Says Claude Helped With A Major Math Problem on ThorstenMeyerAI.com

Buying for a business?Offer from Amazon

Get business pricing on office and shipping supplies

  • Business-only prices and quantity discounts
  • Tax-exempt purchasing
  • Multiple users, one account, clear invoices
As an affiliate, we earn on qualifying purchases.

TL;DR

Anthropic mathematician Levent Alpöge says he worked with Claude to prove the Riemann hypothesis and the broader Generalized Riemann Hypothesis. He released a short note and a large Lean proof archive, but the proof has not been accepted; key verification and expert review remain outstanding.

Levent Alpöge, a mathematician who works at Anthropic, says he and Claude have found a proof of the Riemann hypothesis, a longstanding problem in mathematics. In a note posted November 20, Alpöge also claims the approach proves the broader Generalized Riemann Hypothesis, but neither result has been accepted, and mathematicians are still checking the computer-verifiable proof he released.

Alpöge’s four-page note says he “believe[s]” he has discovered a proof “working with Claude.” It links to a Lean archive divided into 14 parts. Lean is software used to check formal mathematical arguments. The note itself provides a short sketch and a timeline, not a readable presentation of the full proof. Alpöge says the complete human-readable version would exceed 250,000 pages, and that he chose to publish only the formalization.

The size of the archive has been described as 17.4 million lines of Lean code, according to computer scientist Scott Aaronson. MIT mathematician Andrew Sutherland told Scientific American that the code compiled independently “with only the three standard axioms.” That is an encouraging initial check, but it does not by itself establish that the formalized argument proves the intended mathematical statement. Kevin Buzzard said a “comparator run” was still awaited; that check is intended to test whether the code proves the claimed result without relying on an unintended substitute.

Alpöge’s account assigns Claude a substantial role in the technical work. He says he set the direction and guided the search, but that “Claude far outpaced me near the finish line.” He points to Claude’s suggestion about the length of a chain of mathematical constructions called theta lifts. The note does not identify which Claude model was used. Anthropic had not announced the claim, and, according to the source report, had not responded to Nature’s request for comment at the time of publication.

At a glance
reportWhen: Claim posted November 20, 2026; status…
The developmentLevent Alpöge posted a claim that Claude helped him prove the Riemann hypothesis, alongside a Lean code archive that mathematicians are still evaluating.

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

A Claim With Major Mathematical Stakes

The Riemann hypothesis is one of the most prominent unsolved problems in mathematics. It concerns the locations of certain zeros of the Riemann zeta function and has deep connections to how prime numbers are distributed. Many results in number theory are known conditionally, assuming the hypothesis is true; a valid proof could establish those consequences without that assumption and affect a wide range of mathematical work.

The claim also draws attention to a different question: whether an AI system can contribute to research at the frontier of mathematics. Alpöge’s description suggests Claude helped develop technical ideas, but the public evidence so far is his account and an enormous formal code archive. Until independent mathematicians can establish what the code proves and understand the reasoning, the episode is not evidence that the problem has been solved.

The Clay Mathematics Institute lists the Riemann hypothesis among its Millennium Prize Problems, each associated with a $1 million prize. A posted claim does not start the prize process. Clay requires a solution to be published in a refereed outlet, undergo at least two years of scrutiny, and gain general acceptance among mathematicians.

Products Worth Considering

Dell XPS 13 13.4-Inch Touchscreen Laptop, Intel Core 5 320, 16GB RAM, 512GB SSD

Dell XPS 13 13.4-Inch Touchscreen Laptop, Intel Core 5 320, 16GB RAM, 512GB SSD

FLEXISPOT 36-Inch Height-Adjustable Standing Desk Converter with Keyboard Tray

FLEXISPOT 36-Inch Height-Adjustable Standing Desk Converter with Keyboard Tray

As an affiliate, we earn on qualifying purchases.

From Riemann’s 1859 Conjecture to Lean

Bernhard Riemann introduced the hypothesis in 1859. In broad terms, it says that all the “non-obvious” zeros of the zeta function have real part equal to one-half. The Clay Mathematics Institute says the hypothesis has been checked for the first 10 trillion solutions. Those computations are evidence about the cases tested, not proof of a statement about infinitely many zeros.

Formal proof assistants such as Lean let mathematicians encode definitions and logical steps so software can check whether each step follows under the stated rules. Formal verification can help catch errors, but a successful compilation does not automatically show that an argument corresponds to the intended theorem. That distinction is particularly relevant here because the archive is exceptionally large and the readable proof has not been released.

Early reactions reflect that gap. Fields Medalist James Maynard told Nature, “If this fully checks out, it is clearly a historic milestone in mathematics.” Terence Tao described it as an “(enormous) Lean-only proof,” while Andrew Granville called the released note “garbled.” These comments do not constitute a verdict on the proof; they point to the need for checking and an explanation that specialists can scrutinize.

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

— Levent Alpöge, in his note

What the Proof Checks Have Established

The central question remains unresolved: whether the Lean archive proves the Riemann hypothesis and the Generalized Riemann Hypothesis as stated. The reported compilation check is not the same as independent confirmation of the theorem. The comparator run was still pending in the source material, and no complete, accessible written proof has been made available for conventional expert review.

It is also not clear how much of the mathematical strategy and technical work came from Alpöge, Claude, or other tools. The available account comes from Alpöge, and the model used is not identified. The source report says Anthropic had not responded to a request for comment; it also reports no journal publication or acceptance by a mathematical body. The claim should therefore remain attributed to Alpöge rather than described as a proven result.

Independent Review Is the Next Test

The immediate next step is further technical checking of the Lean archive, including the pending comparator run and scrutiny by specialists in the relevant mathematics and formal verification. Researchers will need to determine both whether the code compiles under appropriate assumptions and whether its formalized theorem matches the Riemann hypothesis Alpöge says it proves.

Any claim that survives those checks would still require mathematical review and publication. Under Clay’s rules, the prize would be considered only after publication in a refereed journal, at least two years of scrutiny, and general acceptance. No timetable for those steps has been given, and the available material does not say when Alpöge may publish a fuller exposition or respond to questions about the proof.

Key Questions

Has the Riemann hypothesis been proved?

Not according to the information available. Alpöge has claimed a proof, but no journal or mathematical body has accepted it, and independent checks are ongoing.

What role does Alpöge say Claude played?

Alpöge says he guided the strategy and that Claude contributed substantially to the technical work, including suggesting a chain length for theta lifts. That description comes from Alpöge; the model used and the precise division of work have not been independently confirmed.

What is the Lean archive?

Lean is software that checks formalized mathematical arguments. Alpöge released the proof as a 14-part archive. A reported compilation is an initial technical check, not by itself confirmation that the archive proves the intended theorem.

Does the claim qualify for the Clay prize?

Not at this stage. Clay’s rules require publication in a refereed journal, at least two years of scrutiny, and general acceptance by mathematicians before a solution can qualify.

Source: ThorstenMeyerAI.com

This content is for general information only and is not financial, tax or legal advice. Consult a qualified professional for decisions about your money.
You May Also Like

Decoding AI: How Training Shapes Responses

Exploring how AI models are trained and how their training influences their responses, behavior, and limitations in real-time interactions.

AI Automation Network Security Checklist: What To Review

A hardware-focused checklist outlines what to verify for AI and automation networks, from firewalls and VPN routing to backups and UPS power.

AI Breakthroughs To Watch In 2026: The Top 10

A comprehensive look at the most significant AI advancements expected in 2026, highlighting confirmed developments and ongoing research efforts.

What Is A ‘Weight Virus’ And What Does The Paper Say?

A HCLTech paper tests weight-space misfolding in one small model, but does not demonstrate a self-spreading AI virus or offensive capability.