OpenAI宣布其AI模型已解决千禧年大奖问题中的纳维-斯托克斯存在性与光滑性问题，该项目运行约10,000个并发Agent，花费数百万美元1。然而这一声明随即陷入学术诚信风波——纽约大学数学家Tristan Buckmaster指控OpenAI未恰当引用其与Anthropic员工Levent Alpöge的相关研究1。
Buckmaster发文称OpenAI员工给出两个选项：要么由OpenAI先发表论文次日跟进，要么与OpenAI合作但排除Alpöge参与1。当被询问OpenAI是否获得其工作转录本时，OpenAI否认了该指控，但对是否基于这些转录本训练模型的问题未作回应1。
Buckmaster和Alpöge在周一发布了简化版纳维-斯托克斯方程的证明，他们近一年来使用OpenAI和Anthropic的公开模型进行研究1。两支团队都采用了数学家Diego Córdoba和Luis Martínez-Zoroa开创的方法论1。
这一事件引发了更广泛的学术担忧——UCLA数学家Terence Tao在Mastodon发文警告，AI公司过早且不透明地解决问题可能对数学整体进步产生负面影响1。该争议反映出AI公司在数学研究中日益占据主导地位，可能改变整个数学领域的未来1。
OpenAI announced that its AI model had solved the Navier-Stokes existence and smoothness problem, one of the Millennium Prize Problems, deploying approximately 10,000 agents in parallel and spending millions of dollars on the effort 1. However, the claim quickly became mired in controversy over academic attribution and research ethics.
On Monday, mathematicians Tristan Buckmaster from New York University and Levent Alpöge from Anthropic released their own proof of a simplified version of the Navier-Stokes equation, work they had been developing over the past year using publicly available models from OpenAI and Anthropic 1. Buckmaster subsequently disclosed that OpenAI staff had presented him with two options: either Buckmaster's team would publish first with OpenAI following the next day, or the groups would collaborate on a paper that excluded Alpöge 1. Buckmaster further questioned whether OpenAI had obtained transcripts of his team's research and whether the company's models had been trained on such materials; OpenAI denied having received transcripts but did not respond to the second question 1.
Both research teams drew upon methodologies pioneered by mathematicians Diego Córdoba and Luis Martínez-Zoroa 1. UCLA mathematician Terence Tao warned that AI companies resolving problems prematurely and without transparency could have detrimental effects on mathematical progress as a whole 1. The dispute underscores the growing dominance of AI firms in mathematics research and raises questions about how advances in the field will be attributed and developed in the future.
评论
还没有评论，欢迎留下第一条。