Photo via The Guardian

Why I won’t leave mathematics, despite AI solving centuries-old problems | Ken Ono

Read the original at The Guardian ↗
Cite

Ono, K. (2026, November 22). Why I won’t leave mathematics, despite AI solving centuries-old problems | Ken Ono. The Guardian. https://www.theguardian.com/commentisfree/2026/nov/22/why-i-wont-leave-mathematics-despite-ai-solving-centuries-old-problems

Dated from the outlet's own feed. Check the original for its byline.

The Guardian · collected 2026-11-22 · by Ken Ono

Quick Summary

Ken Ono, a mathematician, reflects on his near decision to quit mathematics following the announcement that AI had solved the Riemann hypothesis, a 167-year-old problem. Ono led the number theory team for Epoch AI’s FrontierMath project, where he witnessed an unreleased ChatGPT model perform complex calculations with ease. Despite these challenges, Ono decided to continue his work in mathematics, inspired by his father's perspective that AI is reminiscent of Ramanujan’s unexpected discoveries. Ono now works at Axiom Math, using AI to formalize and verify mathematical proofs, including the formalization of the classification of finite simple groups, which involved 43.1 million lines of Lean code and led to the discovery of a problem in a published proof.
Written by AI on 2026-11-22, from this article's own text only (the story summary below draws on all the coverage).
See how they differ ↓

AI analysis

Story summary

A mathematician at Anthropic, Levent Alpöge, claims to have used the company’s AI chatbot Claude to prove the Riemann Hypothesis, one of the most challenging problems in mathematics that has remained unsolved for over 160 years. Alpöge, who previously worked with a New York University professor on another famously difficult problem, says Claude made critical breakthroughs that surpassed his own expertise. The Riemann Hypothesis is one of seven Millennium Prize Problems selected by the Clay Mathematics Institute in 2000, offering a prize of US$1 million for a solution. To validate the proof, Alpöge released the answer on his website in a programming language called Lean, allowing computers to verify the solution, which is reportedly 250,000 pages long.

Written for “Anthropic AI Solves Riemann Hypothesis” on 2026-11-22, grounded in this article and the 12 other(s) covering the same event.

How they differ

11 outlets covered “Anthropic AI Solves Riemann Hypothesis”.

What each outlet says

One AI summary per outlet of its latest article on this story, written from that article alone and checked against it. Open the article for the outlet's own words.

The GuardianThis outlet22 Nov

Ken Ono, a mathematician, reflects on his near decision to quit mathematics following the announcement that AI had solved the Riemann hypothesis, a 167-year-old problem. Ono led the number theory team for Epoch AI’s FrontierMath project, where he witnessed an unreleased ChatGPT model perform complex calculations with ease. Despite these challenges, Ono decided to continue his work in mathematics, inspired by his…

Why I won’t leave mathematics, despite AI solving centuries-old problems | Ken Ono (this article)2 articles from The Guardian on this story
Daily Mail20 Nov

Dr. Levent Alpöge, a 34-year-old Turkish-American mathematician, claims to have proven the 167-year-old Riemann Hypothesis using Anthropic's AI chatbot Claude. The Riemann Hypothesis, introduced in 1859 by Bernhard Riemann, concerns the distribution of prime numbers along the number line. Alpöge published his proof in Lean computer code, and experts will now verify its correctness. Clay Mathematics offered a $1…

Most famous unsolved maths riddle solved by AI giant Anthropic after 167 years →2 articles from Daily Mail on this story
The Hindu20 Nov

Levent Alpoge, a mathematician at Anthropic, announced on November 20, 2026, that he had used the company’s Claude chatbot to prove the Riemann Hypothesis, a 160-year-old mathematical problem. Alpoge claims to have also solved the broader Generalized Riemann Hypothesis, but the proof has yet to be reviewed by independent experts. The Riemann Hypothesis is one of the seven Millennium Prize Problems, each carrying a…

Anthropic researcher says Claude solved maths’ greatest riddle →
The Straits Times20 Nov

On November 20, Levent Alpoge, a researcher at Anthropic, claimed that the company’s AI chatbot Claude helped prove the Riemann Hypothesis, a 160-year-old mathematical problem. Alpoge posted a four-page note online but the claim has not been verified by experts. The Riemann Hypothesis is one of seven Millennium Prize Problems, each carrying a $1 million reward if a solution is published in a peer-reviewed journal…

Anthropic researcher says AI chatbot Claude solved math’s greatest riddle →
New York Post20 Nov

Levent Alpöge, a mathematician formerly associated with New York University, claims that an AI named Claude made significant progress on the Generalized Riemann Hypothesis, a complex mathematical problem worth $1 million from the Clay Mathematics Institute. Alpöge used the computer program Lean to verify the proofs generated by Claude, but the work is so extensive that Alpöge himself cannot fully comprehend it…

Anthropic researcher claims Claude cracked legendary $1M math problem — and left him struggling to keep up →
BBC20 Nov

An Anthropic employee and mathematician, Levent Alpöge, claims to have used the company's AI chatbot Claude to prove the Riemann hypothesis, a nearly 170-year-old unsolved math problem related to prime numbers. Alpöge made this claim on Friday, a day after Anthropic's stock market debut, stating that he worked on the proof over six weeks. If verified, this proof would be a significant breakthrough in mathematics…

In a year of AI solving maths problems, Anthropic may have just solved the biggest →
Show 5 more outlets
Toronto Star21 Nov

Levent Alpöge, a mathematician at Anthropic, claims to have solved the Riemann Hypothesis with assistance from the company’s AI tool Claude. Announced on Friday, this development follows a similar claim by OpenAI regarding the Navier-Stokes equation in September. The Riemann Hypothesis, a 150-year-old unsolved problem concerning the distribution of prime numbers, carries a $1 million prize from the Clay Mathematics…

Anthropic worker claims to have solved a Holy Grail of math problems with the help of AI →
Deutsche Welle21 Nov

Levent Alpoge, a mathematician at Anthropic, claims to have used the company’s AI tool, Claude, to solve the Riemann hypothesis, a famous unsolved problem in mathematics dating back to 1859. Alpoge announced on his website that he and Claude generated over 250,000 pages of mathematical material, though the proof still needs verification by other mathematicians. The Riemann hypothesis, one of the Clay Mathematics…

Anthropic researcher uses AI to solve math's greatest riddle →
NPR21 Nov

Levent Alpöge, a mathematician at Anthropic, claims to have solved the Riemann Hypothesis, a long-standing mathematical problem, with the assistance of the company's AI tool, Claude. Alpöge posted a four-page explanation on his website, detailing his discovery which he believes is proven through 17 million lines of Lean code. The claim is significant as the Riemann Hypothesis, formulated in 1859, remains one of the…

Anthropic worker claims to have solved a Holy Grail of math problems with the help of AI →
Al Jazeera21 Nov

A mathematician from Anthropic, Levent Alpoge, claims to have proven the Riemann Hypothesis, a 167-year-old mathematics problem, with the help of an AI system named Claude. Alpoge posted a four-page explanation on his website but did not publish a conventional paper; instead, he used Lean, a programming language and proof assistant, to verify his reasoning. Independent experts are attempting to validate the proof…

Riemann hypothesis: Has the world’s most famous maths problem been cracked? →

Where they agree and split

Framing

How the first headlines put it

A claim to check4
Attributed to a source2
Stated as fact2
The headlines (8)

Anthropic researcher claims Claude cracked legendary $1M math problem — and left him struggling to keep up New York Post · a claim to check

Claude has solved most famous problem in mathematics just a day after Anthropic's record stock market listing, scientist claims Daily Mail · a claim to check

Anthropic worker claims to have solved a Holy Grail of math problems with the help of AI Toronto Star · a claim to check

Anthropic worker claims to have solved a Holy Grail of math problems with the help of AI NPR · a claim to check

Anthropic researcher says Claude solved maths’ greatest riddle The Hindu · attributed to a source

Anthropic researcher says AI chatbot Claude solved math’s greatest riddle The Straits Times · attributed to a source

In a year of AI solving maths problems, Anthropic may have just solved the biggest BBC · stated as fact

Anthropic researcher uses AI to solve math's greatest riddle Deutsche Welle · stated as fact

The full story: what happened, how it changed, and all 13 articles →

Compared with similar articles

This article reads unscored and hedges 7% of its claims. Each row says how that neighbour differs.
Toronto Star
Hedges 3 of 23 claims Outlet record 65/100

Story

Anthropic AI Solves Riemann Hypothesis
Science · 13 article(s) covering the same event. Compare every outlet's account →
How this article was measured signals, leaning charts, publisher record, byline, topics and claims

Signals How these are calculated →

Claims extracted
61
sentences that state something as happening or true
Uncertain
7%
4 of 61 hedged ("reportedly", "may", "according to")
Leaning
not political
takes no side on a contested political question
Publisher signals
63/100
The Guardian's corrections and hedging, across everything collected from it. Not a measure of accuracy.
Outlets on this story
13
Science
Same framing elsewhere
1
articles that frame the event the way this one does
Analyzed 2026-11-22 · how these are computed

How this is being covered How these are calculated →

Article leaning vs. publisher reliability
Source leaning vs. consistency

Publisher

The Guardian · 2676 article(s) · 10 correction(s) detected
Running correction rate · 10 correction(s)

Who wrote this

Ken Ono
1 article(s) here · 1 carrying a prediction
In May 2025, I led the number theory team for Epoch AI’s FrontierMath project.
The only article under this byline in the corpus.

Topics

ChatGPT Epoch AI’s FrontierMath Lean Mathlib

Subjects

Ramanujan PERSON · 5× ChatGPT ORG · 1× Claude PERSON · 1× Epoch AI’s ORG · 1× Hindu NORP · 1× Levent Alpöge PERSON · 1× Manjul Bhargava PERSON · 1× Riemann PERSON · 1× Roger Heath-Brown PERSON · 1× Tamagawa PERSON · 1×

Narrative

The declaration A Severe Misalignment of AI in Mathematics, signed by Terry Tao and other Fields Medalists, warns about rushed announcements, inadequate attribution, and the loss of mathematical understanding.
framing: assertive · carried by 1 article(s) · first seen 2026-11-22
In May 2025, I led the number theory team for Epoch AI’s FrontierMath project.

Claims (61 extracted, 4 hedged)

Last year, I nearly quit mathematics. asserted
I → quit → mathematics
On 20 November, Levent Alpöge announced that, working with Claude, he had proved the Riemann hypothesis, a 167-year-old problem about the distribution of prime numbers. asserted
he → announce → numbers
The Lean and Mathlib communities have since certified the formal proof. asserted
communities → certify → proof
We are only beginning to understand the argument. asserted
We → begin → argument
What kind of future awaits someone beginning a life in mathematics? asserted
kind → await → mathematics
In May 2025, I led the number theory team for Epoch AI’s FrontierMath project. uncertain
I → lead → project
We wrote difficult problems to test large language models. asserted
We → write → models
I watched an unreleased ChatGPT model compute all the moments of Tamagawa numbers of elliptic curves, drawing on work by Manjul Bhargava and Roger Heath-Brown, and on papers my father wrote in the 1960s. asserted
father → watch → 1960s
I was stunned. asserted
I → stun → ?
I earned my PhD in 1993 and built a life around research, teaching, and mentoring. asserted
I → earn → research
I’d love to prove that Ramanujan’s tau function never vanishes, or that half of the partition numbers are even. asserted
half → love → numbers
My 25-year-old self would have been crushed by what we are witnessing today. asserted
we → crush → what
Nothing I write can reassure someone just starting a career. asserted
I → write → career
I can explain why I stayed. asserted
I → explain → ?
My father, a number theorist, died in January. asserted
father → die → January
As a boy, I learned from him about Ramanujan, whose notebooks have occupied mathematicians for generations. asserted
notebooks → learn → generations
Ramanujan believed a Hindu goddess gave him mathematical formulas in his dreams. asserted
goddess → believe → dreams
When I considered working on AI, I feared my father would see it as abandoning mathematics. asserted
father → consider → mathematics
Instead, he told me AI reminded him of Ramanujan: discoveries from an unexpected source might leave us much to understand. uncertain
discoveries → tell → much
I do not take that comparison literally. asserted
I → take → comparison
My father’s response brought me to tears. asserted
response → bring → tears
It gave me permission to keep being a mathematician in a way I had not anticipated. asserted
I → give → way
Today, I am on leave from the University of Virginia and work as Founding Mathematician at Axiom Math. asserted
I → work → Math
We use AI to formalize, verify and extend mathematical proofs in Lean, a language in which a small checking program verifies each step. asserted
program → use → step
This work can expose gaps and make proofs inspectable and reusable. asserted
proofs → expose → gaps
Human judgment remains essential: we must check that the formal statement says what we intended. asserted
we → remain → what
We released the first complete Lean formalization of the classification of finite simple groups, one of mathematics’ largest proofs. asserted
We → release → proofs
Announced in 1983, the classification required decades of work by many mathematicians; a major gap was filled in 2004. asserted
gap → announce → 2004
Our AI helped produce 43.1 million lines of Lean code. asserted
AI → help → code
In the process, we found a problem in a published route to a key theorem and supplied an alternative argument. asserted
we → find → argument
Lean’s kernel checks the resulting formal proof. asserted
kernel → check → proof
The mathematics belongs to the generations who developed it. asserted
who → belong → it
Our formalizations of the classification and other bodies of math are already helping us build and check further mathematics. asserted
us → help → mathematics
After 24 days of work, I watched the final milestones land on my laptop. asserted
milestones → watch → laptop
I cried. asserted
I → cry → ?
I thought about the people who had made the classification possible, and about the father I could no longer tell. uncertain
I → think → father
I’ve advised 35 PhD students, 17 postdocs, and hundreds of undergraduates. asserted
I → advise → undergraduates
I’ve had opportunities young mathematicians cannot assume. asserted
mathematicians → have → opportunities
I cannot ask a graduate student to regard uncertainty as an adventure because I have found rewarding work within it. asserted
I → ask → it
At a Berkeley colloquium two days after OpenAI announced its Navier–Stokes solution, a scheduled lecture became a town hall lasting almost two and a half hours. asserted
lecture → announce → solution
…and 21 more, not listed.
Give feedback · no sign-in needed

No sign-in needed for now. Your note is sent without a name, so it cannot be answered.

History Support