# A machine that proved the Riemann hypothesis, or at least we think it did (slim)

> Vlad Cealicu on Levent Alpöge's note of 20 November 2026 saying he and Claude proved the Riemann hypothesis and its generalised form, posted as 17.4 million lines of Lean with no paper. Two people outside Anthropic compiled it the same day and nobody has read it. The file that states the result is 16 lines and a checker reads the rest, which makes it another win for AI verification. Nothing in it breaks RSA or Bitcoin, and the comparator checks, the model and the Clay prize are still open.

- Full: https://www.anchorterminal.com/blog/a-machine-that-proves-riemann.md (~4,900 tokens) · this version ~2,480 tokens · JSON https://www.anchorterminal.com/blog/a-machine-that-proves-riemann.json · canonical https://www.anchorterminal.com/blog/a-machine-that-proves-riemann
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-11-21
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)

By Vlad Cealicu*, 20 November 2026. 2,066 words, about 9 min to read.

The Riemann hypothesis has been proved by a mathematician working with a language model, or at least we think it has been.

For anyone who hasn't met it, Bernhard Riemann wrote the conjecture down in 1859. It says where a function called the zeta function equals zero, and people care because those zeros govern how the prime numbers thin out as you count upwards [[7]](#ref-7).

- 16: lines in the file that states what was proved
- 17,374,215: lines of Lean in the proof, in 175,187 files
- 41 days: from the first dated step in the note's timeline, 6 October, to the entry that reads GRH, 16 November
- 250,000+: pages the argument came to when Alpöge had Claude write it out for people, which he didn't post

The proof as it was posted on 20 November 2026, from the note [1] and the archive [2].

## What "we think" rests on

Andrew Sutherland at MIT compiled the whole archive first, shortly before noon.

Nobody has read the proof, Alpöge included.

## The 16 lines

We downloaded the archive on Friday night, 14 files that join into one, to see what it says was proved [[2]](#ref-2).

The file that says what was proved is 16 lines long, the main file is 12, and this is the main file's first theorem [[2]](#ref-2).

`RiemannHypothesis` there comes from Mathlib, the Lean community's shared library, at the revision the project pins, and it's three lines [[11]](#ref-11).

In words, every zero of the zeta function that isn't one of the trivial ones has real part one half.

Lines of Lean a person has to read, against the lines only the checker reads

| Part | Lines |
| --- | --- |
| Mathlib's definition of the hypothesis | 3 |
| The main file, five theorems | 12 |
| The file that states what was proved | 16 |
| Everything else in the archive | 17374187 |

Each series has a zero-based scale.

So what a person has to read to know what's being claimed is a few lines, plus the definitions in Mathlib those lines are built on.

## Why this problem fell

On 7 November I wrote about METR's report on how fast Anthropic is improving its own models, and I said I thought [the acceleration was real and narrower than it sounds](/blog/a-machine-that-builds-the-next-machine).

I still think the first half of that is right.

The second half I'm less sure of than I was on 7 November.

2026, from two thirds of the zeros to all of them

1. **10 August** (anthropic). An unreleased Claude raises the share of zeros proved to be on the line from 41.6% to 67.2%, and Alpöge is one of two staff mathematicians who check it. Anthropic writes that it doesn't expect those techniques to lead to a proof of the hypothesis [[6]](#ref-6)
2. **30 September** (openai). OpenAI's first proof of the quasi-Riemann hypothesis, every zero between 1/8 and 7/8, checked in Lean [[7]](#ref-7)
3. **6 October** (openai). OpenAI announces 377 results in one day, the quasi-Riemann proof among them [[7]](#ref-7)
4. **6 October** (anthropic). The earliest date in Alpöge's timeline, on four of its twenty steps [[1]](#ref-1)
5. **15 November** (anthropic). Eight of the twenty steps carry this date, a Sunday [[1]](#ref-1)
6. **16 November** (anthropic). The last step reads GRH [[1]](#ref-1)
7. **20 November** (anthropic). The note and the Lean go up, and Sutherland compiles it before noon [[1]](#ref-1) [[4]](#ref-4)
8. **20 November** (by hand). Ravi Vakil posts Alessandro Masullo's 940-page proof of the same hypothesis, written without AI [[4]](#ref-4) [[5]](#ref-5)

## The rest of Friday

Tao's post lists two more results with Alpöge's name on them, both with Buckmaster of New York University and both with Lean attached.

The same day Ravi Vakil, the president of the American Mathematical Society, posted a 940-page manuscript by Alessandro Masullo that claims the generalised hypothesis by hand [[4]](#ref-4).

## Does it break anything?

I spent twelve years building a crypto data company, so the question I'd expect first is whether this breaks anything, and it doesn't (your coins are as safe as they were on Thursday, however safe that was).

What the generalised hypothesis gave computing was mostly a guarantee about testing whether a number is prime.

## What it does to mathematicians

Alpöge calls it the loss of a guiding light as well as a dawn [[1]](#ref-1).

## What I'd copy

We grade the tools agents use from public evidence and put the source next to every score, and our lab asks open-weight models questions about this site that have known answers and [publishes how they did](/lab/evaluation/).

If you're building with agents, write the statement yourself, whether it's a test, a schema or 16 lines of Lean, and make everything the machine writes pass it.

## What's still open

The comparator runs hadn't finished on Friday evening [[4]](#ref-4).

Anthropic hadn't announced anything itself when Quanta published [[4]](#ref-4).

The Clay prize needs the proof published in a qualifying outlet, two years to pass after that and general acceptance among mathematicians [[8]](#ref-8), so it's at least two years away (and I have no idea who they'd make the cheque out to).

I'm a paying customer of Anthropic, the research agents and the review panel behind this site run on Claude, and this post was edited with Claude's help, which is what the star next to my name is for, so weigh what I say about a Claude result with that in mind.

I think it's been proved, it's another win for AI verification and I think that is splendid, and I hope to see a lot more of this.

---

More in [Opinion](https://www.anchorterminal.com/blog/opinion/index.md) · Older opinion [OpenAI's agents now finish half its two-to-four-hour research tasks](https://www.anchorterminal.com/blog/openai-agents-half-the-research.md)

- Older: [What to build that lasts. The jobs agents ask for that nobody finishes](https://www.anchorterminal.com/blog/the-jobs-nobody-finishes.md)
- Newer: [I asked an agent to book Castle Howard, and the only thing it couldn't do was pay](https://www.anchorterminal.com/blog/i-asked-an-agent-to-book-castle-howard.md)
- All posts: https://www.anchorterminal.com/blog/index.md

Anchor Terminal grades the tools AI agents run on, with a source for every score.

## References

Sources read on 20 November 2026. The archive was downloaded and counted the same evening.

[1] Levent Alpöge, untitled four-page note, 20 November 2026. Source for the claim in his words, the extension to the Generalized Riemann Hypothesis, the formalisation, his not having mastered the mathematics around the argument, the 250,000 pages and why he posted only Lean, who chose the direction and where Claude went past him, the step he credits to Claude, the timeline of twenty dated steps from 6 October to 16 November 2026, the loss and the dawn, the timing and the constraint on announcing, the people he thanks and his expectation that the argument will be simplified.

[2] Anchor Terminal's count of the Lean archive the note links, lean.tar.gz in 14 parts, downloaded on 20 November 2026 at 22:33 UTC.

[3] Terence Tao, Unforced Euler singularity, unforced Navier-Stokes singularity, and the Generalized Riemann Hypothesis, 20 November 2026. Source for the three results in order of how digested they are, the internal model, the Langlands program and arithmetic geometry, the time the digesting will take, and the state of the Euler and Navier-Stokes papers.

[4] Quanta Magazine, Transformation, Konstantin Kakaes, the entries of 19 and 20 November 2026. Quanta reports the compiles by Andrew Sutherland and Simon DeDeo, the variant of Lean, the comparator checks still running and how long they might take, DeDeo's and Maksym Radziwiłł's comments, the 940-page Masullo manuscript and his earlier proof, the Lean community's discussion of formalising it, Anthropic having made no official announcement, its reading of the timing as the IPO, elliptic curves in Bitcoin and web browsing, and Dartmouth not admitting graduate students in mathematics for the next academic year.

[5] Ravi Vakil, Hurried Announcement: Alessandro Masullo GRH, 20 November 2026. Source for the proof being written without AI, Brian Conrad's months of checking and the plan to polish it for a year.

[6] Anthropic, Claude has improved on a longstanding lower bound for the fraction of zeros of the Riemann zeta function that satisfy the Riemann hypothesis, 10 August 2026. Source for 1859 and the million-dollar prize, the bound going from 41.6% to 67.2%, Alpöge and Ralph Furman checking it, and Anthropic not expecting those techniques to prove the hypothesis.

[7] Quanta Magazine, AI Proved the Quasi-Riemann Hypothesis.

[8] Clay Mathematics Institute, Rules for the Millennium Prize Problems, adopted 26 September 2018. Source for the three conditions before a proposed solution is considered.

[9] Wikipedia, Generalized Riemann hypothesis, as read on 20 November 2026. Source for Miller's 1976 result on the Miller-Rabin test under the hypothesis and the unconditional test of Agrawal, Kayal and Saxena in 2002. https://en.wikipedia.org/wiki/Generalized_Riemann_hypothesis

[10] Levent Alpöge and Tristan Buckmaster, Finite-time blow-up for the Navier-Stokes equations, repository README, 20 November 2026. Source for the statement files, the solution file and the comparator configs that allow only the axioms propext, Quot.sound and Classical.choice.

[11] Mathlib, Mathlib/NumberTheory/LSeries/RiemannZeta.lean at revision 0df444a, the one the archive's lakefile pins.
