Anthropic公司数学家莱文特·阿尔波格(Levent Alpöge)于当地时间11月20日公开宣称,他与人工智能模型Claude合作完成了对黎曼猜想的证明 23。这一成果被视为近期AI在解决著名数学问题上超越人类的最新案例,若获确认将可能重塑数学研究领域 1。阿尔波格表示:“我相信,我与 Claude 合作找到了黎曼猜想的证明。” 2
据报道,该证明过程历时约六周,不仅涵盖了经典的黎曼猜想,还涉及范围更广的“广义黎曼猜想” 23。为了让数学界能够逐步验证推理的正确性,阿尔波格将证明编写为Lean代码并发布 23。黎曼猜想由伯恩哈德·黎曼于1859年提出,主要涉及素数的分布规律,是克雷数学研究所设立的千禧年大奖难题之一,每道此类难题提供100万美元奖金 3。目前,其他数学家已开始对其提供的Lean代码进行验证工作 3。
尽管取得了潜在突破,但该声明仍需经过严格的独立验证 1。值得注意的是,就在阿尔波格发布声明的同一天,意大利数学家亚历山德罗·马苏洛(Alessandro Masullo)也发布了一份未使用AI辅助、但尚未完成的黎曼猜想证明 3。此外,此事引发了关于私人科技公司投入巨资利用AI解决数学难题的伦理争议及对环境成本影响的担忧 3。华威大学数学家詹姆斯·罗宾逊批评科技巨头“烧钱展示模型”的行为是不成熟的炫耀,并指出了其潜在的生态损害 3。今年10月,OpenAI公布的一批揭示质数分布规律的数学研究成果也被视为此类技术应用的早期突破口 2。
Levent Alpöge, a Turkish-American mathematician at Anthropic, has claimed that he and the company’s artificial intelligence model, Claude, have proven the Riemann hypothesis. 23 The announcement was made on November 20, with Alpöge stating, "I believe I found a proof of the Riemann hypothesis by collaborating with Claude." 2 According to his report, the work extended beyond the original conjecture proposed by Bernhard Riemann in 1859 to also cover the broader Generalized Riemann Hypothesis. 23 To facilitate verification, the proof was encoded in Lean, a programming language designed for checking mathematical reasoning step-by-step. 23 Other mathematicians have begun reviewing the submitted code, though the claims remain subject to independent validation. 13 This development is cited as part of a recent trend where AI systems are demonstrating capabilities in solving famous mathematical problems previously thought to be exclusively human domains. 1
The claim has sparked debate regarding the role of private technology companies in fundamental science research. 3 It follows earlier reports from October in which OpenAI highlighted progress in understanding prime number distribution using a budget of $15 million and 10,000 AI systems, a move that drew criticism for its resource intensity. 3 James Robinson, a mathematician at the University of Warwick, characterized such large-scale corporate efforts as immature displays of model capability, raising concerns about environmental costs. 3 In a coincidental development on the same day, Italian mathematician Alessandro Masullo released an alternative, unfinished proof of the Riemann hypothesis that did not involve AI tools. 3 The potential resolution of the problem carries significant weight, as the Clay Mathematics Institute offers a $1 million prize for each solved Millennium Prize Problem. 3
评论
还没有评论,欢迎留下第一条。