San Francisco‑based AI firm Anthropic made headlines on Friday when Levent Alpöge, a member of the company’s technical staff, posted a four‑page explanation claiming that the firm’s AI tool Claude helped him generate a proof of the Riemann Hypothesis. The hypothesis, first formulated in 1859, describes the distribution of prime numbers and has remained one of the seven Millennium Prize Problems identified by the Clay Mathematics Institute.
What Alpöge says
Alpöge wrote that after months of collaboration with Claude, the AI “far outpaced” him near the finish line. He posted the claim on his personal website, noting that the proof was encoded in roughly 17 million lines of Lean code, a programming language used to formalize mathematical proofs. According to Alpöge, Claude ran the code and confirmed that the proof was valid, though he cautioned that he still needed to verify the underlying mathematics himself.
Expert reaction
Alex Kontorovich, a distinguished professor and department chair of mathematics at Rutgers University, explained that normally a proof is presented in a paper that mathematicians can read and scrutinize line by line. He said, “When somebody gives you a paper, that’s also kind of code for a proof and you have to actually read the paper, follow it yourself, figure out if you believe every single line.” Kontorovich added that the computer’s verification is a promising first step, but human understanding remains essential.
Institutional response
The Clay Mathematics Institute, which listed the Riemann Hypothesis as an unsolved problem and offers a $1 million prize for a correct solution, still shows the hypothesis as open on its website. Institute president Martin Bridson described Alpöge’s announcement as “historic in significance” but warned that the path from a claimed proof to broad acceptance will be “long and difficult.”
Context within AI breakthroughs
This claim follows a recent announcement from OpenAI that its technology contributed to a breakthrough on the Navier‑Stokes equations, another long‑standing mathematical challenge. Both instances illustrate a growing trend of AI systems being used as research assistants in highly technical fields.
Anthropic’s market debut
Anthropic went public just days before Alpöge’s post, raising a record‑breaking $81 billion at a $1.5 trillion valuation. CEO Dario Amodei has spoken publicly about the potential for AI to accelerate scientific discovery while also emphasizing the need for responsible development and oversight.
Next steps
Alpöge said he will continue to refine the proof and intends to submit a formal paper for peer review. The mathematics community will likely spend months, if not years, evaluating the claim, testing the Lean code, and determining whether the proof holds up under rigorous scrutiny.
For now, the Riemann Hypothesis remains listed as unsolved, and the broader scientific world watches closely as AI tools like Claude become increasingly involved in frontier research.
Original reporting: KOAT Albuquerque — read the source article.