OpenAI has announced a new independent advisory panel of mathematicians to guide its dealings with mathematical research and the wider mathematics community, including how new results are presented and released. The move follows a string of spectacular mathematical results that turned into a reputational crisis for the company. The panel's abrupt arrival caught many mathematicians by surprise, and researchers told The Verge the group is a good first step while leaving basic questions about its actual function, its influence, and whether OpenAI will listen to it. There are also concerns over whether such a small group of prominent researchers can represent the wider mathematical community.
The stakes are unusually high because OpenAI has spent the last few years planting flags across increasingly difficult terrain in mathematics, and this week it claimed one of its biggest prizes yet: a solution to a legendary Millennium Prize problem. Under normal circumstances that would have been celebrated as a historic achievement. Instead, many mathematicians have watched the company's relentless advance with growing unease, seeing it less as an enthusiastic newcomer than an impossibly well-resourced interloper charging into problems they have dedicated their lives to. At the heart of that unease, according to researchers, is a sense that OpenAI is doing mathematics for different reasons: mathematicians want to advance the field, while OpenAI wants to win.
The company's latest attempt to repair fractured relations is to consult elite practitioners, but mathematicians speaking to The Verge, including one of the group's members, describe a messy and confusing affair bearing many of the hallmarks of OpenAI's previous rushed forays into mathematics. That suggests the company has learned little from its mistakes, the researchers said. The panel also faces a daunting first task: helping coordinate the release of scores more results OpenAI says its unreleased model has produced.
The prospect of that release is already stirring dread among researchers over what a looming tidal wave of breakthroughs could do to their field. Mathematician James Maynard, a University of Oxford professor and winner of the prestigious Fields Medal, told The Verge he has spent a lot of time this past year soul searching, grappling with the future of his field as the traditionally slow-moving discipline hurries to adapt to AI. Days before he spoke with The Verge, OpenAI revealed it had produced solutions to 10 long-standing mathematics problems, some of which had confounded academics for decades.
Separately, another researcher is challenging OpenAI about the data behind its mathematical discoveries. Andreas Thom raised concerns in a series of posts on Mastodon that interactions he and his colleagues had with the ChatGPT chatbot before OpenAI's triumphant announcement may have contributed to its success in the field. One of the 10 results OpenAI announced with great fanfare last month involved Thom's area of expertise, so-called non-sofic groups, and OpenAI acknowledged that the result built heavily on previous work by Thom and fellow mathematician Gábor Kun.
OpenAI's announcement that it solved one of mathematics' legendary Millennium Prize problems should have been a moment of triumph. The result is both an undeniable achievement and a striking demonstration of how rapidly AI is transforming mathematics, but before it was formally announced the breakthrough was complicated by the circumstances that prompted OpenAI to pursue the problem. After hearing other researchers were making progress, it appears to have thrown its considerable resources into a last-minute effort to beat them to the punch, and the ensuing controversy has surfaced allegations of scooping, spying, and flagrant violations of long-standing academic norms that researchers fear could have a chilling effect on the field. Abhishek Saha, a mathematics professor at Queen Mary University of London, described the conduct as the kind of things mathematicians will generally not do.
OpenAI says its solution to the Navier-Stokes problem, which relates to the flow of liquid and gas, came from an internal AI model more powerful than the newly released GPT-6 Astra alongside 10,000 concurrent agents. Navier-Stokes is one of seven Millennium Prize Problems, each carrying a $1 million reward. OpenAI says it started training the internal model on August 28th, and that it has exhibited unprecedented performance in the company's benchmarks, including mathematics. The result is a big breakthrough for the mathematics community, but it does not come without controversy.
More company and startup news from TechManNews.
