Back to News

Anthropic Mathematician Claims Claude-Assisted Proof of Riemann Hypothesis

#riemann-hypothesis#claude#lean#formal-proof

Anthropic mathematician Levent Alpöge claims to have proved the Riemann hypothesis with assistance from Claude, posting a 4-page PDF and Lean formalization code split across 14 compressed files. The proof, not yet independently verified, reportedly extends to the Generalized Riemann Hypothesis and potentially the Extended Riemann Hypothesis. MIT's Andrew Sutherland has published verification records showing all 175,187 modules compiled successfully.

Coverage timeline

  1. Techmeme

    Joseph Howlett / Scientific American : Anthropic mathematician Levent Alpöge claims he used Claude to prove the 167-year-old Riemann hypothesis and posts the proof in the Lean programming language — An AI-generated solution of the 167-year-old problem, long considered far out of reach to human mathematicians …

  2. 量子位量子位

    黎曼猜想被证明?!Claude加持,34岁数学家放出Lean证明 衡宇 2026-11-21 00:07:39 来源: 量子位 衡宇 发自 凹非寺 量子位 | 公众号 QbitAI 刚刚,34岁的Anthropic数学家Levent Alpöge放出一份PDF文档, 声称自己与Claude合作,找到了黎曼猜想 (数学界最有名、悬赏100万美元的百年难题) 的证明。 Alpöge没有发表完整论文,选择公布了整套Lean形式化证明代码,分成14个压缩文件供外界下载。 如果把它转成一份包含全部细节的文档,篇幅会超过25万页。 Alpöge还在文档里表示, 研究到了最后阶段,Claude已经远远跑在了自己前面 ,连他本人都表示自己“在尚未掌握证明周边大量数学知识的情况下,就已经向外界展示了完整论证”。 原文的形容颇为生动: No, Claude flew, and danced, and sparkled near that finish line. 临近答案的终局,Claude不再缓步求索,它振翅腾空,以起舞完成这场奔赴。 按照Levent Alpöge的说法,这套方法还能够解决广义黎曼猜想 (GRH) ,并有望延伸到扩展黎曼猜想。 从1859年提出至今,黎曼猜想这道题已经悬而未决167年。多少数学家前赴后继,至今没人能交出完整答卷。 一场可能改写数学史的AI研究,正在等待验证。 (黎曼猜想是啥?可戳这篇文章速览:《百年黎曼猜想被Claude破了新纪录》) Claude给出了关键思路 Levent Alpöge给出的名为《Remark》的PDF仅有4页。 具体到解决黎曼猜想的数学方法上,这次证明思路借鉴了函数域中解决黎曼猜想类比命题的Stepanov方法。 在函数域里,数学家可以通过构造具有高重数零点的辅助函数推进证明。但将这套方法迁移到普通整数领域时,遇到了一个难题——整数没有与函数求导完全对应的自然结构。 Levent Alpöge在文档里表示,自己引导双方找到了一个合适的数学情境,之后Claude开始展现出令他惊叹的技术能力。 期间,Claude提出了一个新的、关键的建议。 通过连续的Theta Lifts (Theta提升) ,在自守形式框架下逐步累积所需要的重数。 按照文档的描述,每一次提升都能增加一个单位的重数,最终形成一条足够长的提升链。 Claude 还提

  3. Simon Willison

    I believe I have, working with Claude, discovered a proof of the Riemann Hypothesis. In fact, the method easily extends much further, resolving the so-called Generalized Riemann Hypothesis, and I expect, with more time, the Extended Riemann Hypothesis too. Claude and I have also formalized the arguments without assumptions. It is natural to feel that this is an end, in my case a loss of a principal guiding light, but what I saw during the course of progress so amazed me that I believe it is one dawn across science of something far greater. [...] Claude and I have had a great time together with such breadth in recent months, but this time Claude far outpaced me near the finish line. No, Claude flew, and danced, and sparkled near that finish line. — Levent Alpöge , alpo.ge/remark.pdf Tags: anthropic , claude , generative-ai , mathematics , ai , llms , levent-alpoge

  4. The Verge AI

    Anthropic went public on November 19th, 2026. | Photo: Getty Images Anthropic mathematician Levent Alpöge claims he has proved the Riemann hypothesis, one of mathematics's most notoriously difficult problems, with the help of Claude, cracking a puzzle that has eluded some of the world's best mathematicians for decades. If verified, it would mark yet another seismic achievement for AI in mathematics and add to the upheaval already unleashed by OpenAI's previous, often inelegant forays into the field. Alpöge's solution has yet to be verified independently. Anthropic has not commented or involved itself in Alpöge's announcement. In a document on his website , Alpöge says he texted Anthropic CEO Dario Amodei abo … Read the full story at The Verge.

  5. 机器之心机器之心

    太炸了! 刚刚,Anthropic 数学家 Levent Alpöge 在自己的个人网站上上传了一份 PDF 文件,该文件仅有 4 页,但却给出了一个足以载入史册的重磅成果——其宣称在 Claude 的帮助下,著名的 黎曼猜想(Riemann Hypothesis) 已经得到了证明,并给出了 Lean 代码。 报告地址:https://alpo.ge/remark.pdf Lean 证明:https://alpo.ge/lean.tar.gz.part01 直到 https://alpo.ge/lean.tar.gz.part14 此文件一公布便举世皆惊。毕竟这个数学问题自 1859 年提出以来,已经困扰了人类 167 年。 还不仅如此,Alpöge 在文件中也宣称已经解决了 广义黎曼猜想(Generalized Riemann Hypothesis,GRH) ,并有望在投入更多时间后,进一步处理他所称的 Extended Riemann Hypothesis 。 与消息的分量形成反差的是,眼前 这份 PDF 只有 4 页 。它包含成果声明、方法线索、致谢和研究时间线。真正承载完整论证的材料,被放在脚注指向的 14 个 Lean 压缩分卷 中, 整个文稿有 25 万页 。 Alpöge 称,形式化工作已经完成,论证不依赖额外猜想。他同时坦言,自己尚未掌握这份庞大论证周围的全部数学,希望继续理解、简化,并把它写成可供人阅读的文本。 目前可以确认的是,这份证明声明及其材料入口已经出现;本文尚未取得完整工程通过独立核验的结果。从这里到「黎曼猜想已被解决」,还需要数学家检查形式化命题、证明依赖与验证结果。 消息刚出,就已经引起了整个 AI 圈的震动,𝕏与朋友圈已经开始迅速传播蔓延。 至于是否可靠,让消息飞一会吧! 为何一个关于零点的命题如此重要 黎曼猜想关乎素数的分布。2、3、5、7、11……这些只能被 1 和自身整除的整数,是算术的基本构件,但它们出现的间隔并不整齐。 数学家已经能够描述素数总体上怎样变稀疏,黎曼猜想进一步约束的是:实际分布能偏离这个平均趋势多远。克雷数学研究所的介绍,也将「平均分布」与「偏离平均的程度」作为理解这一问题的入口。 可以把它想成估算一条长路沿途的路灯数量:知道平均每公里有多少盏,还不足以判断某一段究竟会比预期多

  6. 机器之心机器之心

    编辑|Panda 11 月 20 日,Anthropic 数学家 Levent Alpöge 宣称, 与 Claude 合作发现了黎曼猜想的证明 ,并把论证以 Lean 代码公开。此前还停在「三分之二零点」与「准黎曼猜想」的进展表,一夜之间又得重写。 https://alpo.ge/remark.pdf 随后,MIT 数学家 Andrew Sutherland 公布核验记录:全部 175187 个模块编译通过,两套独立编写的内核接受了完整导出证明,核心定理与独立目标之间也完成了一种实现的命题比对。 https://math.mit.edu/~drew/grh/ 变化还不止这一道。同一天,Alpöge 与 Tristan Buckmaster 发布无外力纳维–斯托克斯方程的有限时间爆破论文。杨–米尔斯的存在性证明材料则已在 11 月公开,独立小组也给出了首份评估。 https://github.com/tristanbuckmaster/navier-stokes 之前的 10 月 26 日,OpenAI 还宣布了三项由内部模型完成的成果,其中一项是一篇署名只有「OpenAI」的论文,标题直截了当:指数时间假说是错的(The Exponential Time Hypothesis is False)。 https://github.com/openai/math/blob/main/preprints/The-Exponential-Time-Hypothesis-is-False-October-25-2026/paper.pdf 几个小时后,Anthropic 的 GitHub 仓库 formal-math 里多了一个名为 eth 的文件夹。里面是一篇 49 页的论文《A 2^o(n) Time Algorithm for 3-SAT》,作者栏只有:Claude,Anthropic。 https://github.com/anthropics/formal-math/blob/16a5ce41304fcdaea4e0be7c132e1c361dc26da5/eth/paper.pdf 据了解,其核心算法由一个内部 Claude 研究模型在 10 月 16 日找到,输出了 1400 万 token,没有任何人类输入;之后 Anthropic 研究

Related stories

  • Mathematician Alessandro Masullo Publishes 940-Page Human-Written Proof of Generalized Riemann Hypothesis

    Stanford professor and AMS president Ravi Vakil announced that mathematician Alessandro Masullo has publicly released a 940-page proof of the Generalized Riemann Hypothesis, titled 'The Music of the Primes'. The work, developed over nearly 12 years without any AI assistance, was reviewed by Masullo's doctoral advisor Brian Conrad over several months. The team decided to release the current version early due to the current mathematical community atmosphere, urging mathematicians to verify and formalize it.

    1 reports · 1 sources
  • NYU and Anthropic researchers publish AI-aided Navier-Stokes blowup proof, claim to close OpenAI loophole

    NYU professor Tristan Buckmaster and Anthropic employee Levent Alpöge published an AI-aided proof of Navier-Stokes blowup, claiming it closes a loophole in OpenAI's earlier solution. The work comes two months after OpenAI's initial result.

    1 reports · 1 sources
  • OpenAI claims disproof of Exponential Time Hypothesis, plus math and physics results

    OpenAI announced three new discoveries in mathematics, computer science, and physics, including a claimed disproof of the Exponential Time Hypothesis (ETH), a conjecture widely believed for nearly 30 years. The results cover local smoothing for the Euclidean wave equation, the Kakeya conjecture in four dimensions, and a rigorous mathematical existence proof for four-dimensional quantum field theories including quantum chromodynamics. External experts guided the results, and materials were released on GitHub for verification.

    5 reports · 5 sources1153 points · 701 comments
  • Anthropic launches Claude Fable 5.2 and Mythos 5.2 with coding, reasoning, vision gains

    Anthropic released Claude Fable 5.2 and Claude Mythos 5.2, flagship models aimed at complex technical work, with claimed improvements in coding, reasoning, vision, and communication. The models are priced at $8 per million input tokens and $40 per million output tokens, a 20% cut from 5.1, with a fast mode at $16/$80. During prerelease testing, the models formalized a proof of Fermat's Last Theorem in 30 hours, solved two open black hole physics problems, and scored 64.9% on the RoboDojo Sim benchmark, up from 27.6% for Fable 5.1.

    14 reports · 7 sources916 points · 1051 comments

News updates every hour. Subscribe in your RSS reader to get new stories automatically.

Subscribe