After a series of high-profile mathematical announcements spiraled into a reputational crisis, OpenAI has convened an independent panel of elite mathematicians to guide how it and other AI companies engage with mathematical research and its community, according to a report published Monday by The Verge. The nine-member Advisory Group on Mathematics and Artificial Intelligence (AGMAI), hosted by Princeton's Institute for Advanced Study, will advise on how new mathematical results are presented and released. The company disclosed that its unreleased model has now resolved more than 100 long-standing open problems across most areas of mathematics.
The panel comprises mathematicians from Stanford, Harvard, Oxford, and Cambridge, with multiple Fields Medals and MacArthur "genius grants" among them, researchers told The Verge. OpenAI said the group will operate independently, with freedom to offer unsolicited advice, publicly comment on the company's impact on mathematics, and make its recommendations public. Members will receive no payment from OpenAI and the group can change its membership as it sees fit. The group's bare-bones website indicates it "came together after OpenAI approached some of its members about establishing an external advisory board," though which members were initially contacted and how the remaining researchers were selected remains unclear. The group's formation caught many mathematicians by surprise, with researchers questioning what it will actually do, how much influence it will wield, and whether such a small cadre of prominent academics can represent the wider mathematical community.
Advisory group member Martin Hairer, a professor at Imperial College London and EPFL, wrote in a Tuesday blog post that "it is a fact that AI companies have in recent months been producing some high profile mathematical results and that their publication and dissemination has been falling far short of acceptable mathematical practice." He stressed that the group is "genuinely independent of OpenAI and any other of the so-called 'frontier' AI labs" and that members have not signed restrictive agreements beyond obvious confidentiality requirements. "The other fundamental fact is that our overarching aim is to represent as well as we possibly can the interests of the mathematical community," Hairer wrote, acknowledging the group is hardly a representative sample but insisting all nine members "take this extremely seriously and are acting in good faith."
The effort acknowledges what the report describes as a monumental PR disaster, in which OpenAI faced accusations of scooping researchers, failing to credit the human work its models built upon, and even stealing from those using its models. These controversies left many mathematicians wondering if the company understood or respected the discipline it had entered. The small panel size raises questions about whether the group has the breadth of expertise needed to evaluate everything coming its way, given the highly specialized skills required at the frontiers of mathematics. Francesco Fournier-Facio, an incoming mathematics professor at Heriot-Watt University, described the panel as feeling "like an ivory tower," while Simon Machado, a researcher at ETH Zurich, questioned whether elite mathematicians understand the concerns of the broader community. "They're amazing mathematicians, but I don't know if they are the people I want to represent me in more political questions," Machado said, noting that the reality facing such elite positions differs vastly from what most mathematicians face right now.
The group's immediate priority is advising OpenAI on how to coordinate the release of the large number of significant results its internal model has reportedly produced, according to its website. Whether the panel can earn mathematicians' trust will determine how effective it becomes at containing fallout from recent missteps and softening the blow for whatever the company plans to announce next. Kevin Buzzard, a mathematics professor at Imperial College London, questioned why OpenAI needed such an elite body to understand what the community wants after so much has been written on the topic: "It wasn't proofs of hard theorems, it was better understanding of our subject." The formation of this advisory structure suggests AI companies may need to balance competitive pressures with academic norms that have governed research communities for generations, even as the speed of algorithmic discovery outpaces traditional peer review cycles.