25位菲尔兹奖得主联合发表公开信，指责OpenAI等AI实验室在解决数学问题时存在学术不诚实行为1。这些指控包括仓促发表未经验证的证明、未能正确署名合作者，以及可能利用他人研究成果等问题1。纽约大学教授特里斯坦·巴克马斯特声称，OpenAI曾向他施压，要求他不要在Anthropic合作者的署名中予以认可1。
数学家们对问题的严重性提出了明确关切。根据公开信，"这些解决方案通常匆匆宣布，没有留出适当的著述空间、新方法和想法的隔离，以及对他人以往相关工作的引用"1。OpenAI的相关证明迄今仍未获得验证1。作为回应，OpenAI本周已撤销对加州理工学院一项数学活动的赞助1。
Twenty-five Fields Medal recipients have issued a public letter accusing OpenAI and other artificial intelligence laboratories of academic misconduct in their approaches to solving renowned mathematical problems 1. The mathematicians allege that these organizations have rushed to publish unverified proofs, failed to properly credit collaborators, and potentially misappropriated others' research 1.
Among the signatories' specific complaints, New York University professor Tristan Buckmaster has accused OpenAI of pressuring him not to credit Anthropic collaborators on joint work 1. The letter emphasizes that solutions are frequently announced hastily, "leaving no time for a proper writeup, the isolation of new methods and ideas, and citing relevant previous work of others" 1. OpenAI's mathematical proofs remain unverified 1. The controversy reflects growing concern within the mathematics community that intensifying competition among AI firms threatens established academic norms and practices 1.
In a significant escalation, OpenAI withdrew its sponsorship of a mathematics program at the California Institute of Technology this week 1. The dispute builds on concerns previously raised through the Leiden Declaration, a statement released in June by a working group of mathematicians 1.
评论
还没有评论，欢迎留下第一条。