{"schema":"sanctity-signal/0.2","record_type":"signal","id":"alpoge-grh-lean-proof-e107","domain":"science","drop_id":"signal-e107","occurred_at":"2026-11-20","how_to_use":"This is a structured record of one real-world action and its competing interpretations, from Sanctity Signal. To build content from it (a reel script, a blog, a thread): open on the hook, ground the viewer with plain and observed, walk the dated trail, present the interpretations as live tension between named readings, and close on the before/after shift and what_to_watch. Hard rules: items under established are fact; items under unknown are never presented as fact; the evidence status (verified or reported) is stated, not dropped; sources are cited.","actor":"Levent Alpöge","action_type":"publication of a 4 page claim and a 175,187 module Lean archive asserting a proof of the Riemann Hypothesis, written with Claude, then checked by outside kernels","observed":"Alpöge's note at alpo.ge/remark.pdf opens: \"I believe I have, working with Claude, discovered a proof of the Riemann Hypothesis.\" It runs 4 pages, undated, first person, with a timeline from 10/6/26 to 11/16/26, the last entry the generalized result. He writes that he and Claude \"formalized the arguments without assumptions\", that Claude \"far outpaced me near the finish line\". He states he released Lean only, that the informal argument would exceed 250,000 pages printed, that he is presenting a complete argument \"before having mastered a massive neighbourhood around its mathematics\". Andrew Sutherland of the Massachusetts Institute of Technology published his own checks at math.mit.edu/~drew/grh/, updated 22:05 Universal Time November 20. His record: all 175,187 modules compile with zero errors on Lean v4.33.1, commit 819816b2, against Mathlib tag v4.33.1, commit 0df444a3, the build finishing at 16:43:21. Printing axioms for the 5 theorems in Main.lean returns only propext, Classical.choice and Quot.sound. The nanoda kernel exited clean at 19:05:45. con-leche accepted 1,468,418 declarations in verified mode at 21:19:04, and its compare mode accepted 51,692 challenge declarations with all 5 theorems \"justified by the solution\" at 21:19:12. leanchecker replayed 175,187 of 175,187 modules with no rejections at about 18:17; lean4lean passed the same count at about 21:55. Sutherland records that the end-to-end comparator on Lean v4.36.0 was still running, that the statement comparison and axiom check \"already passed in one run\", and that the default and paranoid export kernels had returned no verdict. His caveats: the build ran unsandboxed and called lean directly rather than through lake build, with only a 77 module sample matched against a real lake build. His page also notes that the generalized claim covers Dirichlet L-functions only. Also on November 20, Ravi Vakil of Stanford, president of the American Mathematical Society, posted a competing all-human draft by Alessandro Masullo, wrote \"It is all human, with no AI\", and 8 hours later said several people had reported problems in it.","corrections":[],"evidence":{"status":"verified","note":null,"sources":[{"role":"primary","label":"Levent Alpöge, remark.pdf at alpo.ge: the 4 page note and Lean archive claiming the proof, posted Nov 20, 2026","url":"https://alpo.ge/remark.pdf"},{"role":"primary","label":"Andrew Sutherland, MIT: independent build, kernel replays and axiom audit, updated 22:05 UTC Nov 20, 2026","url":"https://math.mit.edu/~drew/grh/"},{"role":"corrob","label":"The Decoder: Stanford releases an all-human competing draft hours after the Claude proof claim, Nov 21, 2026","url":"https://the-decoder.com/hours-after-anthropic-claims-a-claude-proof-of-the-riemann-hypothesis-stanford-releases-an-all-human-competing-draft/"},{"role":"corrob","label":"Scientific American via Yahoo: independent mathematicians attest the proof, Nov 20, 2026","url":"https://www.yahoo.com/news/science/articles/ai-riemann-hypothesis-proof-almost-210000837.html"},{"role":"corrob","label":"riemann-rev: independent reading project dashboard of who has checked what, record Nov 20, 2026","url":"https://github.com/leecrossley/riemann-rev"}]},"strength":{"cost":"High","commitment":"High","reversibility":"Low","persistence":"Structural","rationale":"Cost is high: the archive is the product of 6 weeks of work by a working number theorist and a frontier model, published under his own name with his reputation attached. Commitment is high, because a named claim on the most watched problem in mathematics cannot be walked back quietly. Reversibility is low in the same sense: the files are public, pinned to a commit, and anyone can rebuild them. Persistence is structural rather than dated. Whether or not this proof survives, the method is now demonstrated, and the next formal claim will arrive the same way."},"shift":{"prior":"AI helps mathematicians with steps, and people write the proofs.","action":"Alpöge's note at alpo.ge/remark.pdf opens: \"I believe I have, working with Claude, discovered a proof of the Riemann Hypothesis.\" It runs 4 pages, undated, first person, with a timeline from 10/6/26 to 11/16/26, the last entry the generalized result. He writes that he and Claude \"formalized the arguments without assumptions\", that Claude \"far outpaced me near the finish line\". He states he released Lean only, that the informal argument would exceed 250,000 pages printed, that he is presenting a complete argument \"before having mastered a massive neighbourhood around its mathematics\". Andrew Sutherland of the Massachusetts Institute of Technology published his own checks at math.mit.edu/~drew/grh/, updated 22:05 Universal Time November 20. His record: all 175,187 modules compile with zero errors on Lean v4.33.1, commit 819816b2, against Mathlib tag v4.33.1, commit 0df444a3, the build finishing at 16:43:21. Printing axioms for the 5 theorems in Main.lean returns only propext, Classical.choice and Quot.sound. The nanoda kernel exited clean at 19:05:45. con-leche accepted 1,468,418 declarations in verified mode at 21:19:04, and its compare mode accepted 51,692 challenge declarations with all 5 theorems \"justified by the solution\" at 21:19:12. leanchecker replayed 175,187 of 175,187 modules with no rejections at about 18:17; lean4lean passed the same count at about 21:55. Sutherland records that the end-to-end comparator on Lean v4.36.0 was still running, that the statement comparison and axiom check \"already passed in one run\", and that the default and paranoid export kernels had returned no verdict. His caveats: the build ran unsandboxed and called lean directly rather than through lake build, with only a 77 module sample matched against a real lake build. His page also notes that the generalized claim covers Dirichlet L-functions only. Also on November 20, Ravi Vakil of Stanford, president of the American Mathematical Society, posted a competing all-human draft by Alessandro Masullo, wrote \"It is all human, with no AI\", and 8 hours later said several people had reported problems in it.","update":"A machine can carry a famous proof that no person has read."},"inference":{"established":["Alpöge published a 4 page note and a Lean archive on November 20.","All 175,187 modules compile; 3 outside kernels accepted the export.","The 5 main theorems use only Lean's 3 standard axioms."],"unknown":["That the Lean statement matches the Riemann Hypothesis as mathematicians state it.","That the proof is accepted. No readable write-up or referee report exists.","How much Claude contributed. Prompts and transcripts are not published."]},"narrative":{"hook":"An Anthropic mathematician published a claimed proof of the Riemann Hypothesis. Outside kernels accept it.","plain":"Levent Alpöge posted a 4 page note and 175,187 Lean modules at alpo.ge, written with Claude. Caveat: no readable proof exists, and the comparator run was unfinished.","story":["One of the 7 Millennium Prize Problems may now be settled in a form no mathematician can read. Lean checks every step mechanically, so correctness no longer needs a human reader. What peer review is for, and who gets credit, are now open questions."],"trail":[{"d":"SEP 4","t":"Anthropic published a computer-checked proof of Fermat's Last Theorem that its model wrote in 11 days. Edition 013 carded it. Edition 034 carded a physics calculation Claude finished on about $100 of computing."},{"d":"NOV 16","t":"The last entry in Alpöge's own timeline, and the one he marks as the generalized result. He says he contacted Anthropic's chief executive for clearance before announcing."},{"d":"NOV 20","t":"Alpöge posts the 4 page note and the Lean archive at alpo.ge. He publishes no readable mathematics, saying the informal argument would run past 250,000 printed pages."},{"d":"NOV 20","t":"Sutherland compiles all 175,187 modules with zero errors, replays them through 3 further kernels, and finds only the 3 standard axioms. The end-to-end comparator is still running at his last update, 22:05 Universal Time."},{"d":"NOV 20","t":"Vakil posts Masullo's all-human competing draft, then updates 8 hours later to say several people reported problems in it. Two claimed proofs of the same statement are now in the open."},{"d":"Reacts","r":true,"t":"Number theorists, who must audit a statement no one has read in prose; the Clay Mathematics Institute, which holds a $1 million prize with a publication rule; the Lean and Mathlib maintainers; journal editors, who have no process for a 175,187 module submission; Anthropic, whose model is named throughout, 2 days after its listing."}],"glossary":[{"term":"Riemann Hypothesis","def":"An 1859 conjecture about where the zeta function equals zero. It fixes how irregularly prime numbers are spaced, and much of number theory is written assuming it."},{"term":"Lean","def":"A proof assistant. A human or model writes each step as code, and Lean's kernel refuses anything that does not follow. It cannot check that the code states the intended theorem."},{"term":"Standard axioms","def":"The 3 assumptions Lean proofs may use freely. A proof that needs extra axioms is weaker, so listing them is how readers check nothing was smuggled in."},{"term":"Comparator run","def":"A check that the code has not redefined the symbols in its own statement, run against a reference file. It is the test of whether the theorem proved is the theorem claimed."}]},"judgment":{"prompt":"Which explanation do you believe?","interpretations":[{"id":"a","label":"Formal proof changed mathematics","text":"A machine-checked archive settles a 167 year problem with no readable proof.","abstain":false},{"id":"b","label":"One mathematician's unusual method","text":"Alpöge formalizes by habit. A Lean-only release reflects him, not a trend.","abstain":false},{"id":"c","label":"Anthropic showcasing its model","text":"A lab researcher announced 2 days after the listing, with Claude named throughout.","abstain":false},{"id":"d","label":"The normal formalization cycle","text":"Lean proofs of large theorems have landed before. This is the next one.","abstain":false},{"id":"e","label":"Too early to call.","text":"The evidence cannot yet separate these.","abstain":true}],"distribution":{"status":"held back until you have made your call","calls":null}},"story_thread":null,"what_to_watch":[{"linked_interpretation":"a","direction":"strengthens","observable_event":"A completed end-to-end comparator run and a published audit by a named number theorist confirming the Lean statement is the Riemann Hypothesis, by January 31, 2027.","window_end":"2027-01-31","state":"unresolved"},{"linked_interpretation":"a","direction":"weakens","observable_event":"A named mathematician publishes a specific error in the statement or the formalization, or Alpöge withdraws or materially amends the archive, by January 31, 2027.","window_end":"2027-01-31","state":"unresolved"},{"linked_interpretation":"c","direction":"strengthens","observable_event":"Anthropic publishes the prompts, transcripts or compute cost behind the proof, or markets the result in a product or investor communication, by December 31, 2026.","window_end":"2026-12-31","state":"unresolved"}]}