Riemann Hypothesis × Claude

Verification log · living record

Riemann Hypothesis × Claude: verification log

A dated, sourced record of every independent build, code audit, expert assessment and formal paper concerning the claimed proof announced on November 20, 2026. Each entry states when it happened, where it was reported, what the evidence says and which status it changed.

Last entry: 2026-11-20 · Sources are re-checked about every two hours

Current status · Claimed proof · verification in progress · updated Nov 20, 2026
  • Lean archive published

    Posted at alpo.ge on November 20, 2026, in 14 parts, alongside a four-page note.

    Done
  • Independent build of the Lean code

    On launch day, mathematicians posting on X reported that the archive was still compiling, a process expected to take hours. No completed independent build has been reported yet.

    In progress
  • Main theorem states RH and GRH faithfully

    Mathematician Elliot Glazer reported on X that the formalized statement is GRH. A full public audit of the statement and every definition behind it has not been published.

    Preliminary
  • Only standard axioms, no gaps

    No public #print axioms report for the archive has been published yet.

    Pending
  • Human-readable write-up

    Alpöge says he intends to write the argument properly. No paper has been posted.

    Not yet
  • Expert review and broad acceptance

    For a result of this size, understanding and acceptance usually take months or longer.

    Not yet
  • Clay Millennium Prize

    Clay's rules require publication in a qualifying journal, two years, and general acceptance.

    Not yet

Event log, newest first

  1. Community review#2026-11-20-zulip

    Lean experts begin examining the archive on the Lean community Zulip

    Number theorist Thomas Bloom (University of Manchester) pointed readers to the Lean community's Zulip chat, where experts were investigating whether the formal proof is valid. No conclusion had been posted when this entry was written.

    Status change: Independent build: Not started → In progress

    Source: OfficeChai roundup of reactions (citing Thomas Bloom on X) · Secondary report

  2. Statement check#2026-11-20-statement-grh

    Elliot Glazer: the formalized statement is GRH; code still compiling

    Mathematician Elliot Glazer posted on X that the formalized statement in the release is the Generalized Riemann Hypothesis, that the code was still compiling and would take a few hours, and that he expected it to compile. This is a reading of the statement, not a verification of the proof.

    Status change: Statement check: Pending → Preliminary

    Source: Elliot Glazer on X · Primary (individual expert post)

  3. Media report#2026-11-20-sciam

    Scientific American reports the claim

    Scientific American reported Alpöge's claim, noting that the proof was posted as Lean code, that issues can arise in Lean, and that experts should not need long to check whether the hypothesis is coded correctly and the proof is correct.

    Source: Scientific American · Secondary report

  4. Announcement#2026-11-20-announcement

    Levent Alpöge announces a proof of RH and GRH made with Claude

    A four-page note on alpo.ge states that Alpöge believes he has, working with Claude, discovered a proof of the Riemann Hypothesis, that the method resolves the Generalized Riemann Hypothesis, and that the arguments were formalized in Lean without assumptions. The formalization is released as 14 archive parts (lean.tar.gz.part01 to part14). The note says an informal draft exceeded 250,000 pages and was not posted.

    Status change: Lean archive published: — → Done

    Source: Levent Alpöge, statement (PDF) · Primary

  5. Background#2026-08-10-zeta-67

    Earlier result: Claude proves more than two-thirds of zeta zeros lie on the critical line

    Anthropic reported that an unreleased Claude raised the proven share of zeros on the critical line from 41.6% to 67.2%, with a Lean formalization that passes the comparator validation tool, examined by Brian Conrey and Dan Goldston. This is a separate, partial result, not a proof of RH.

    Source: Anthropic research post · Primary

How this log works

This page is maintained as a public fact record, so that articles, students and researchers can point to a single place showing what has and has not been verified about the claimed proof, and when.

  • Event time. Each entry carries the date of the event in UTC. Where the original source gives no exact time, only the date is shown.
  • Original source. Every entry links to the source it relies on. We prefer primary sources: the author's own note, the Lean archive, build logs, posts by the experts themselves, preprints and official statements.
  • Evidence. The entry summarizes what the source actually shows, not what it implies. Claims, readings and verifications are labelled as such.
  • Status change. When an event moves a verification step forward or back, the change is recorded and the tracker at the top of the page is updated in all six languages of this site.

Status definitions

  • Done: the step is complete and the evidence is public.
  • In progress or Preliminary: work has started, or an early assessment exists, but no complete public result yet.
  • Pending or Not yet: nothing public yet.

Corrections

If you have a source that contradicts or adds to an entry, please send it through the About and corrections page. Corrections are logged, never silently overwritten. For background, see Has the Riemann Hypothesis been solved? and the overview of the claimed proof made with Claude.

About this log

What gets added to this log?

Only verifiable events that bear on whether the proof is correct: independent builds of the Lean archive, audits of its statement and axioms, public assessments by mathematicians or Lean experts, statements by the author or institutions, and formal papers or preprints. Commentary that adds no new evidence is left out.

How often is it checked?

Sources are reviewed about every two hours. A new entry is published only when there is genuinely new, verifiable information; quiet periods produce no entries.

How should I cite an entry?

Link to the entry's anchor, for example riemannhypothesisclaude.org/verification-log#2026-11-20-announcement, and cite the original source listed in that entry wherever possible.

What do the source tiers mean?

Primary means the evidence comes directly from the person or body that produced it, such as the author's note, a build log or an expert's own post. Secondary report means it comes from journalism or a roundup that cites the primary source.