RESEARCH · RESEARCH · #3997
Anthropic mathematician Levent Alpöge says he solved the Riemann Hypothesis with Claude
Levent Alpöge, an Anthropic researcher, has posted a claimed proof of the Riemann Hypothesis that he says he developed working with Anthropic’s Claude; the material — including a Lean formalization — is published on his website but has not been independently verified, Anthropic has not publicly commented, and the Clay Mathematics Institute has not accepted the result. Alpöge also asserts the method extends to the Generalized and Extended Riemann Hypotheses.
KEY POINTS
- Levent Alpöge, an Anthropic researcher, has posted a claimed proof of the Riemann Hypothesis that he says he developed working with Anthropic’s Claude; the material — including a Lean formalization — is published on his website but has not been independently verified, Anthropic has not publicly commented, and the Clay Mathematics Institute has not accepted the result.
- Alpöge also asserts the method extends to the Generalized and Extended Riemann Hypotheses.
- If validated, a machine-assisted proof of the Riemann Hypothesis would be a landmark result for both mathematics and the application of AI-assisted formal reasoning, but the claim currently lacks independent verification and formal acceptance.
DECISION BRIEF
CONFIDENCE
WHAT CHANGED
Levent Alpöge (an Anthropic researcher) published a claimed proof of the Riemann Hypothesis that he says he developed working with Anthropic’s Claude. The materials, including a Lean formalization, are posted on his website.
WHY NOW
Alpöge’s claim is public and accompanied by a machine-checkable Lean formalization; if validated, a machine-assisted proof would be a landmark for mathematics and AI-assisted formal reasoning. The claim remains unverified and has been released amid attention to Anthropic (including its recent IPO), increasing scrutiny and rapid community discussion.
WHO IS AFFECTED
Directly affected: the mathematical community (number theory / experts who will verify or refute the proof), Anthropic (employer and developer of Claude), and the Clay Mathematics Institute (which administers the Millennium Prize and must accept solutions). The broader AI-formalization research community is also directly implicated by the claim’s reliance on a model and a Lean formalization.
CONFIRMED
- Levent Alpöge has posted a claimed proof of the Riemann Hypothesis on his website and states he worked with Claude on it. (sources report this directly.) - The posted materials include a Lean formalization. - The claim has not been independently verified. - Anthropic has not publicly commented on the claim in the cited reporting. - The Clay Mathematics Institute has not accepted the result (and, per its rules, acceptance requires publication and scrutiny). - Alpöge asserts the method extends to the Generalized and Extended Riemann Hypotheses (as stated in his post). - Mathematicians are actively discussing and examining the posted material (e.g., community chat/Zulip reactions reported).
UNCERTAIN
Missing or conflicting evidence (explicitly flagged): - Missing: Independent mathematical verification that the proof is correct. Sources say the proof has not been independently verified. - Missing: Formal acceptance by the Clay Mathematics Institute. Sources state it has not been accepted. - Missing: Any public confirmation or detailed comment from Anthropic regarding Claude’s specific role or the company’s stance on the claim; sources report Anthropic has not commented and previously said (in August) it did not expect Claude’s techniques to prove the Riemann Hypothesis (potentially conflicting with a claim that Claude was “integral”). - Unclear: The precise technical dependence on Claude versus human insight—sources quote Alpöge’s statement that he “worked with Claude,” but do not detail the division of labor or which steps, if any, are machine-only versus human-constructed. - Missing / incomplete: TechCrunch excerpt contains a truncated sentence about a collaborator (Tristan Buckmaster) and therefore does not provide full evidence about any additional co-authors or releases. - Missing: Community consensus on correctness; reports note active analysis and some confidence about the Lean formalization but not a settled verdict.
WHAT TO WATCH
Concrete observable signals to monitor (from the cited reporting): - Publication of formal peer reviews, referee reports, or detailed expert analyses from number theorists examining the posted Lean files. - Updates to Alpöge’s website or the Lean repository (corrections, expanded explanations, or additional formalized files). - Any public statement or clarification from Anthropic about Claude’s role or the company’s assessment. - Statements or actions from the Clay Mathematics Institute regarding submission status, verification process, or acceptance. - Ongoing community discussion (e.g., Zulip chat or similar forums) reaching technical summaries or consensus about the argument’s key steps and validity.
WHY IT MATTERS
If validated, a machine-assisted proof of the Riemann Hypothesis would be a landmark result for both mathematics and the application of AI-assisted formal reasoning, but the claim currently lacks independent verification and formal acceptance.
EVIDENCE MAP
4Editorial claims linked to specific sources, with support, contradiction and context shown separately.
Levent Alpöge posted a claimed proof of the Riemann Hypothesis that he says he developed working with Anthropic’s Claude.
SUPPORTEDChecked 2026-11-20 · 2 supporting
The Verge states Alpöge claims he solved the Riemann hypothesis 'with the help of Claude' and quotes his website saying he 'believe[s] I have, working with Claude, discovered a proof'.
SUPPORTS TechCrunch AIMEDIA · 2026-11-20TechCrunch reports Alpöge’s essay describing finding a Lean-compilable solution and credits Claude in his account of discovering the proof.
Alpöge has published the material, including a Lean formalization, on his website.
SUPPORTEDChecked 2026-11-20 · 2 supporting
The Verge reports that Alpöge presented his solution in Lean and that the material has been published on his website.
SUPPORTS TechCrunch AIMEDIA · 2026-11-20TechCrunch notes the proof is Lean-compilable and mentions the formalized result is published on Alpöge’s website.
The claimed proof has not been independently verified and has not been accepted by the Clay Mathematics Institute.
SUPPORTEDChecked 2026-11-20 · 2 supporting
The Verge states Alpöge’s solution has yet to be verified independently and that the Clay Mathematics Institute has not publicly accepted the result.
SUPPORTS TechCrunch AIMEDIA · 2026-11-20TechCrunch says it will be a while before the mathematics community has a handle on the proof and indicates the formalized result needs verification.
Anthropic has not publicly commented on Alpöge’s announcement; earlier the company said it did not expect Claude’s techniques to prove the Riemann Hypothesis.
SUPPORTEDChecked 2026-11-20 · 1 supporting
SOURCES & TIMELINE
2It’s a big week for Levent Alpöge: The mathematician saw his employer, Anthropic, become the first publicly-traded frontier AI lab. He also appears to have solved one of the most important problems in mathematics. “The precise timing is horrible,” he wrote in an essay announcing his findings and describing how he messaged Anthropic CEO Dario Amodei last Sunday night to ask how to announce that he had found a Lean-co…
An Anthropic mathmetician’s latest breakthrough is another sign of how quickly AI is transforming mathematics. If you buy something from a link, The Verge may earn a commission. See our ethics statement. Anthropic mathematician Levent Alpöge claims he has solved the Riemann hypothesis, one of mathematics’ most notoriously difficult problems, with the help of Claude, cracking a puzzle that has eluded some of the wor…