OpenAI Just Claimed a Huge Math Discovery. Some Academics Are Crying Foul
OpenAI Just Claimed a Huge Math Discovery. Some Academics Are Crying Foul
OpenAI 宣称取得重大数学突破，部分学者提出质疑
OpenAI said today that it has found an AI-generated solution to one of the biggest problems in mathematics—a 200-year-old equation that describes the natural behavior of fluids like water and air. OpenAI 今日表示，他们已利用人工智能找到了数学领域最大难题之一的解决方案——这是一个已有 200 年历史的方程，用于描述水和空气等流体的自然行为。
The announcement appears to demonstrate the stunning power for AI to advance mathematics. But it has been marred by claims from another mathematician, Tristan Buckmaster, who says that OpenAI rushed ahead to solve the problem after learning of the work of Buckmaster and a colleague—Anthropic researcher Levent Alpöge. Buckmaster claims OpenAI also tried to influence who got credit for the work. 这一公告似乎展示了人工智能推动数学发展的惊人力量。然而，另一位数学家 Tristan Buckmaster 的指控给这一成就蒙上了阴影。Buckmaster 表示，OpenAI 在得知他和同事——Anthropic 研究员 Levent Alpöge 的研究工作后，便匆忙抢先解决该问题。Buckmaster 还声称，OpenAI 曾试图干预该研究成果的署名权。
The proof concerns the Navier-Stokes equation, one of the unsolved problems in the Clay Millennium prizes, which are each worth $1 million. 该证明涉及纳维-斯托克斯方程（Navier-Stokes equation），这是克雷数学研究所“千禧年大奖难题”之一，每道题的奖金高达 100 万美元。
Sebastien Bubeck, a mathematician and AI researcher at OpenAI, said in a press briefing that the company began training a new AI model with advanced mathematical capabilities on August 28. OpenAI 的数学家兼 AI 研究员 Sebastien Bubeck 在新闻发布会上表示，该公司于 8 月 28 日开始训练一个具备高级数学能力的新型 AI 模型。
After reading rumors that Anthropic was making progress toward solving Navier-Stokes, Bubeck said the company decided to dedicate more resources to tackling the problem. The company had more than 1,000 agents tackle the problem over more than 50 hours; eventually, as many as 10,000 agents were hard at work before the company discovered that it had come up with a solution. 在读到有关 Anthropic 在解决纳维-斯托克斯方程方面取得进展的传闻后，Bubeck 称公司决定投入更多资源攻克这一难题。该公司动用了超过 1,000 个智能体（agents）进行了 50 多个小时的攻关；最终，在发现已得出解决方案之前，有多达 10,000 个智能体在同时工作。
“I thought there must be a mistake somewhere,” Bubeck said. “And on Sunday morning we had the final solution, Lean-formalized and everything.” (Lean is a programming language that can be used to formalize mathematical proofs.) “我当时想，这中间肯定哪里出错了，”Bubeck 说，“到了周日上午，我们得到了最终的解决方案，并且完成了 Lean 形式化验证等所有工作。”（Lean 是一种可用于数学证明形式化的编程语言。）
OpenAI noted that solving this problem required using considerably more computing power than it had previously spent on solving mathematical problems. The amount required cost “in the millions of dollars,” said Mark Chen, head of research at OpenAI. OpenAI 指出，解决这个问题所消耗的计算能力远超其以往在数学问题上投入的资源。OpenAI 研究主管 Mark Chen 表示，所需的计算成本高达“数百万美元”。
On Monday, Alpöge and Buckmaster, a mathematician at NYU, posted documents claiming key advances in an area relevant to the Navier-Stokes problem. The pair says they used several AI models, including Claude and Codex, to complete their work. 周一，Alpöge 和纽约大学数学家 Buckmaster 发布了相关文件，声称在与纳维-斯托克斯问题相关的领域取得了关键进展。两人表示，他们使用了包括 Claude 和 Codex 在内的多种 AI 模型来完成他们的工作。
Buckmaster also posted a statement claiming that, last week, he learned that OpenAI had become aware of his and Alpöge’s work and had started putting significant resources toward the problem. Buckmaster claims that he asked OpenAI leaders about whether the company had accessed the pair’s Codex logs. He says he was told the model “didn’t look up user data” but claims the company didn’t respond to questions about training. He then says that OpenAI offered several “proposals,” including one in which Buckmaster could publish a paper announcing the Navier-Stokes problem had been solved by an internal OpenAI model but without Alpöge’s name included. Buckmaster 还发表声明称，上周他得知 OpenAI 已经获悉他和 Alpöge 的研究，并开始投入大量资源攻克该问题。Buckmaster 声称他曾询问 OpenAI 领导层，公司是否查阅过他们两人的 Codex 日志。他说对方告知该模型“不会查看用户数据”，但他声称公司并未回应有关模型训练的问题。随后他表示，OpenAI 提出了几项“提议”，其中一项是让 Buckmaster 发表论文，宣布纳维-斯托克斯问题已被 OpenAI 的内部模型解决，但论文中不包含 Alpöge 的名字。
In a post on social media, Bubeck pushed back on the idea that the company had suggested removing Alpöge’s name. OpenAI CEO Sam Altman also stepped into the fight, and defended the team’s work. 在社交媒体上，Bubeck 反驳了公司曾建议删除 Alpöge 名字的说法。OpenAI 首席执行官 Sam Altman 也介入了这场争论，并为团队的工作进行了辩护。
Buckmaster, Alpöge, and Anthropic did not immediately respond to WIRED’s request for comment. Buckmaster、Alpöge 和 Anthropic 未能立即回应《连线》（WIRED）杂志的置评请求。
In the briefing, Bubeck and other OpenAI executives denied that the company had ever inspected the pair’s Codex prompts in order to inform their work. “We, whether it’s the researchers or the agents, did not see any of their work until it was released publicly last night,” he said. 在发布会上，Bubeck 和其他 OpenAI 高管否认公司曾为了指导自身研究而查看过两人的 Codex 提示词。“无论是研究人员还是智能体，我们在昨晚他们公开发布成果之前，都没有看到过他们的任何工作内容，”他说。
In a blog post detailing the work, the company wrote, “while unlikely, we cannot rule out that de-identified data derived from their usage of our products helped improve our models.” 在详细介绍该工作的博客文章中，公司写道：“虽然可能性不大，但我们不能排除他们在使用我们产品时产生的去标识化数据，在某种程度上帮助改进了我们的模型。”
“I want to be extremely clear that we recognize the priority of Levent Alpöge and Tristan Buckmaster’s work on unforced Euler, and we have nothing but congratulations to them on this monumental achievement that they have made. To be clear, we did not use their prompt or proof to prompt our models or direct our agents,” said Bubeck. “我想非常明确地表示，我们认可 Levent Alpöge 和 Tristan Buckmaster 在无外力欧拉方程（unforced Euler）方面工作的优先权，对于他们取得的这一里程碑式成就，我们表示衷心的祝贺。需要说明的是，我们没有使用他们的提示词或证明来引导我们的模型或指挥我们的智能体，”Bubeck 说道。
Ven Chandrasekaran, a mathematician at the company, also emphasized that their solution was significantly different in nature from the one produced by OpenAI’s model. 该公司数学家 Ven Chandrasekaran 也强调，他们的解决方案在本质上与 OpenAI 模型生成的方案有显著不同。
The spat over who gets credit for solving one of the biggest problems in math may rumble on for a while. As AI takes on more of the work involved with finding proofs, such fights could perhaps become more common. 关于谁应因解决数学界最大难题之一而获得功劳的争论，可能会持续一段时间。随着人工智能承担更多寻找数学证明的工作，此类争端或许会变得更加普遍。