EXPLORINEER
en

KR — 전자신문

Anthropic Researcher Claims Claude Proved the Riemann Hypothesis

AI2 min read

An Anthropic researcher says he and the company's AI chatbot Claude found a proof of the Riemann Hypothesis, one of the hardest unsolved problems in math. The claim has not yet been checked by outside experts.

What happened

Levent Alpöge, a mathematician who works at the AI company Anthropic, posted a claim on his website. He said that he and Claude found a proof of the Riemann Hypothesis. He shared a 4-page document with the claim. Independent experts have not yet checked the work. Brian Conrey, an expert on this hypothesis, called it a huge moment in history, but he has only read the website post and has asked for the full paper.

Key numbers and facts

  • The Riemann Hypothesis was proposed in 1859 by German mathematician Bernhard Riemann.
  • The Clay Mathematics Institute named it one of seven 'Millennium Problems' in 2000. Each problem has a $1 million prize.
  • Alpöge says the AI work filled more than 250,000 pages.
  • The announcement came one day after Anthropic's record stock market listing (IPO).

Background

The Riemann Hypothesis is about prime numbers, which are numbers that can only be divided by 1 and themselves. The idea is that primes, which look random, follow a hidden pattern tied to a special math function called the Riemann zeta function. Primes are used in RSA encryption, which helps keep online banking and crypto safe. So a proof could affect cryptography, quantum physics, and computer science. To be accepted by the Clay Institute, a proof must be published in a peer-reviewed journal and pass two years of checks. Alpöge first shared a version written in Lean, a programming language that checks math proofs by computer. He also said he solved a wider version called the generalized Riemann Hypothesis. Earlier, in September, OpenAI said its AI had solved a different Millennium Problem about fluid flow, the Navier-Stokes problem. Alpöge had worked on that problem with a New York University mathematician, and he raised credit concerns at that time.

Why it matters

If the proof is correct, it would be one of the biggest events in the history of math, and it would show how much AI can do in research. If it is wrong, it would be another example of AI producing large amounts of hard-to-read or low-quality work. Experts say the next few years of checking will decide which one it is. Alpöge also admitted that the timing of the announcement may look suspicious.

What to watch

  • Whether outside mathematicians check the Lean version and the full paper.
  • Whether the paper is accepted by a peer-reviewed journal.
  • Whether simpler proofs appear, as Brian Conrey expects.
  • How OpenAI and other AI companies respond to this claim and their own earlier results.

Glossary

Anthropic
An AI company that makes the Claude chatbot.
Claude
An AI chatbot made by Anthropic.
Riemann Hypothesis
A famous unsolved math problem about how prime numbers are spread out.
Lean
A programming language that uses computers to check whether a math proof is correct.
Clay Mathematics Institute
A US research group that named seven Millennium Problems and offers a $1 million prize for each one.

Ad

Add 150 words. Learn 10 a day.

Dicter word books send only the words you learn to your review notes.

AITechCrunch

Is selling an AI version of your face to this microdrama startup the future of Hollywood?

Microdrama startup Shortical, founded last January, makes short two-minute romance episodes and is now producing AI-generated content. Its latest series, "Inevitable," stars Israeli actor Aki Avni, who licensed his face to the company to create a digital avatar of himself for the wholly AI-made project, which premiered in Los Angeles. Shortical says the offering is in its early days and has not yet resulted in any signed contracts beyond the Avni deal.

AITechCrunch

Here are the top AI agents that can live in your text messages

A growing number of AI agents can be texted like a normal contact, remembering context, connecting to everyday apps and services, and completing tasks such as scheduling, booking, shopping, and reminders. The article highlights Instinct, which reportedly has a $10 billion valuation after raising $1 billion, along with other options like Ask Proxy, Caddy, Comma, and Fambot, which target personal, work, and family use.

AIThe Register

ChatGPT shrunk down to run on a Super Nintendo

Tech journalist Benj Edwards has released ChatGPT for SNES, a tiny chatbot that runs in an emulator or as a ROM on a Super Nintendo cartridge. It is a 2.2 million-parameter distilled version of OpenAI's GPT-OSS-120b, with an optional fact book drawn from simple English Wikipedia, and Edwards says it knows about 8,000 words and often gets things wrong.