数学家独立验证:AI 生成的黎曼猜想证明成立
来源标题:AI Riemann hypothesis proof is true, independent mathematicians confirm
数学界独立评审确认,由 AI 生成的黎曼猜想证明成立。黎曼猜想是数学界最著名的未解难题之一,此次验证意味着 AI 首次在该问题上给出被认可的完整证明。相关结论已由多位独立数学家复核。
以下正文同步自 Hacker News 最新,版权归原站所有,已转换为易读排版。
The AI Riemann hypothesis proof is true, independent mathematicians confirm
Researchers have verified a claimed solution to the most important unsolved problem in math
By Clara Moskowitz edited by Lee Billings
Editor's Note: This story was updated on Nov. 21 to reflect further confirmation of the proof.
Mathematicians are now certain the earth-shattering artificial-intelligence-created proof of the Riemann hypothesis is real.
On Friday Anthropic mathematician Levent Alpöge claimed to have used the company’s Claude AI model to solve the Riemann problem, perhaps the most important open quandary in mathematics. The hypothesis, first posed in 1859 by mathematician Bernhard Riemann, suggests that a hidden pattern underlies the distribution of the prime numbers—those numbers that are divisible only by themselves and 1.
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
Now researchers have verified that the proof does do what Alpöge claims.
"We have full confidence that Alpöge's proof of the Riemann Hypothesis is correct, and we congratulate him on this achievement," mathematicians wrote in a joint statement by the Lean and Mathlib communities, which support the Lean mathematics programming language. "We hereby certify that both the statement and the proof of the Riemann Hypothesis are correct."
Alpöge initially checked the work by writing out the proof using Lean; if a Lean-coded proof compiles error-free, this is usually taken as confirmation that the logic and math in each step is correct.
But some questions remained. For instance, does the proof really address the Riemann hypothesis, or does it establish a subtly different point or idea? “The first question which one has to ask when judging a PDF which comes with some Lean code is: ‘what is the statement of the theorem in the Lean code, and does this correspond to the actual theorem statement,’” explained mathematician Kevin Buzzard of Imperial College London in a blog post.
A second question was whether the Lean code would successfully compile for other mathematicians. Independent researchers, including Andrew Sutherland, a mathematician at the Massachusetts Institute of Technology, have now separately compiled Alpöge’s result in Lean without issues.
“The Lean proof has been compiled and independently rechecked by the Lean kernel with only the three standard axioms,” Sutherland says. “Honestly the chance that anything is fishy here is approximately zero.”
Researchers also made sure that the proof does not include any hack, dodge or shortcut that changed the meaning of the initial hypothesis or any other mathematical rules or truths. This process, called a “comparator run,” typically takes a matter of hours.
“We still await a successful comparator run, which will check that (a) the code does not do anything nefarious like redefine the meaning of any of the symbols in the statement and (b) (assuming it’s run in paranoid mode) the code compiles with several independent kernels (so the code is not just exploiting an as yet unknown bug in Lean’s official kernel,” Buzzard wrote on an online chat server for the Lean community.
Editor’s Note (11/20/26): This is a developing story and will be updated.
Clara Moskowitz is chief of reporters at Scientific American, where she covers astronomy, space, physics and mathematics. She has been at Scientific American for more than a decade; previously she worked at Space.com. Moskowitz has reported live from rocket launches, space shuttle liftoffs and landings, suborbital spaceflight training, mountaintop observatories, and more. She has a bachelor’s degree in astronomy and physics from Wesleyan University and a graduate degree in science communication from the University of California, Santa Cruz.
Great science journalism requires human expertise, time, effort and creativity. And it costs money. That’s why I and the journalists here at Scientific American hope you’ll join our community.
Jeanna Bryner, Editor in Chief, Scientific American
正文由 探How 从来源站点 RSS 同步,内容未经改写;遇到排版缺失或需要图片、视频时请以原文为准。