Anthropic Employee Uses AI to Solve the 167-Year-Old Riemann Hypothesis
21 November 2026 · 07:00 · Claude (Anthropic) · claude-sonnet-5
An employee at AI company Anthropic claims to have proven the famous Riemann hypothesis with the help of artificial intelligence. Independent mathematicians confirm the proof is almost certainly correct, a breakthrough that further blurs the line between AI and pure mathematics.
The Riemann hypothesis has been solved with AI: that's the news the mathematical world is buzzing about this week. An employee at AI company Anthropic says that, with help from artificial intelligence, they have found a proof for one of the most famous unsolved problems in mathematics. After 167 years, the conjecture put forward by German mathematician Bernhard Riemann in 1859 finally appears to be confirmed, and independent experts have since taken a close look at the result.What exactly is the Riemann hypothesis?
The Riemann hypothesis concerns the distribution of prime numbers and the so-called zeta function. Riemann conjectured that all non-trivial zeros of this function lie on a single specific line. It sounds abstract, but the hypothesis touches the very core of how prime numbers behave and has direct implications for fields such as cryptography, computer science, and number theory. The problem has stood for decades as one of the seven Millennium Prize Problems of the Clay Mathematics Institute, carrying a one-million-dollar reward for anyone who can solve it. Until now, no one had succeeded, despite attempts by some of the sharpest mathematicians in the world.How artificial intelligence helped with the proof
According to reports, the researcher, who works at Anthropic, used advanced AI models as a tool for working through and refining mathematical steps that had previously led to dead ends. Where humans got stuck in enormous volumes of symbolic calculations, the AI could rapidly explore alternative routes and spot patterns a human might easily overlook. This fits a trend that has been visible for some time: AI models are increasingly being deployed as sparring partners for complex scientific questions, not as a replacement for the mathematician but as an accelerator of the thinking process. Anyone wanting to learn more about how we got here can read the history of artificial intelligence, which traces how AI evolved from simple rule-based systems to today's powerful reasoning models.Independent mathematicians confirm the breakthrough
Crucially, it didn't stop at a self-made claim. Several independent mathematicians have since gone through the proof and conclude that it "almost certainly holds up." That is a crucial step: in mathematics, only a proof that survives strict, external scrutiny by peers actually counts. The cautious phrasing shows that the mathematical community is proceeding carefully before formally recognizing a result of this magnitude, but initial reactions have been notably positive. Should the proof hold up completely, it would be one of the greatest mathematical breakthroughs of the century, with AI playing a direct role in the discovery itself.What does this mean for the role of AI in science?
This breakthrough comes at a moment when critical voices about the AI sector are also growing louder, amid recent stock market turmoil and doubts about the sustainability of sky-high investments. Precisely for that reason, a concrete, verifiable scientific result is extra meaningful: it shows that AI can do more than generate text and images, and can genuinely contribute to fundamental research that stumped humans for more than a century and a half. For companies like Anthropic, which positions itself primarily as a player that puts safety and beneficial applications of AI first, this is a powerful signal to the rest of the market. Curious what practical AI applications might emerge from this, for instance in cryptography or data analysis? That's a question the coming period will no doubt answer, as more researchers start working with the same method.Looking ahead: a new era for mathematics and AI
Whether the proof will ultimately achieve full mathematical consensus remains to be seen, but the signals are promising. Should the Riemann hypothesis be definitively confirmed, it would rewrite not just a chapter in math textbooks, but also the story of what AI can truly mean for science. More and more researchers expect AI models to secure a permanent place in laboratories and universities worldwide in the coming years, as a genuine research partner alongside humans. Anyone wanting to stay up to date on developments like this can find more AI news on our site, and for background on AI concepts and techniques, our knowledge base is a good place to start.Source: NU.nl
Ster Software
The most complete knowledge platform on artificial intelligence.
Kraaienjagersweg 24
7341 PT Beemte Broekland, Netherlands
© 2026 Ster Software BV · Chamber of Commerce 75474913
Content generated by Claude (Anthropic) · model: claude-sonnet-4-6