OpenAI Creates Independent Math Panel as It Claims 100+ Problem Solutions

OpenAI on September 21, 2026, announced an independent Advisory Group on Mathematics and Artificial Intelligence, hosted at the Institute for Advanced Study, and said an internal model it began training on August 28, 2026, has resolved more than 100 long-standing open problems across most areas of mathematics.

The Model Claim Behind the Group

In addition to resolving the Navier–Stokes Millennium Prize problem, the internal model has now resolved more than 100 long-standing open problems across most areas of mathematics, OpenAI said. The company said the pace of the model’s progress in mathematics surprised the mathematicians within OpenAI and led to internal discussions about the best way to inform the community of the rapid progress in order to prepare and adapt the field.

OpenAI described mathematics as a fundamental science in which novel discoveries can produce a wide range of applications, and said this makes the promise of these capabilities substantial and their responsible development and deployment important beyond mathematics itself. The company said it recognizes this and is working through how broader deployment of math-related AI capabilities can be done responsibly.

The Open Letter OpenAI Cited

The announcement pointed to an open letter, “A Severe Misalignment of AI in Mathematics,” published September 11, 2026, in which mathematicians raise concerns about the negative externalities of solving open problems as a benchmark for new AI systems. OpenAI said the criticisms highlight the need for thoughtful engagement of AI companies with the mathematical community.

In the letter, the signatories argue that the push by AI companies to solve mathematical problems as a benchmark is detrimental to the science of mathematics and to the mathematical community, writing that “the goals of the AI companies and the goals of the mathematical community are severely misaligned.” The letter says solutions are often announced in a rush, leaving no time for a proper writeup, for the isolation of new methods and ideas, or for citing relevant previous work, which it says raises severe attribution and plagiarism questions. It also warns that the accelerating mass production of true/false statements could destroy fertile ground for new ideas, and that AI-conceived results would remain unrealized without mathematicians willing to develop them, connect them to prior work, and carry them into the accepted body of mathematics.

The letter states that AI offers the potential to enhance and accelerate genuine mathematical study and understanding, but that whether the changes ultimately benefit the field or have a destructive effect will be determined in large part by the decisions of the humans in control of the technology. It says the issues must be addressed urgently by the mathematical community, by the companies developing these technologies, and by a society that will confront similar problems in many other forms of intellectual work. Each listed signatory is identified with a Fields Medal year; the names include Artur Avila, Manjul Bhargava, Simon Donaldson, Martin Hairer, June Huh, Peter Scholze, Terence Tao, and Maryna Viazovska, among others. The letter is registered under DOI 10.5281/zenodo.22737750.

What OpenAI Says the Group Will Do

According to the announcement, the group will advise on the review and communication of emerging results: helping OpenAI assess their significance, advising on how to coordinate their dissemination, and advising on academic and professional standards of mathematical research. It will also advise on how OpenAI’s tools can support mathematical research and learning, and the company said its aim is to give mathematicians capable tools for their own research.

OpenAI said the group will operate independently, with the freedom to offer advice the company has not requested, to comment on OpenAI’s impact on mathematics, and to make its advice public. The company said the group’s value depends on its members applying independent judgment and challenging the company’s own, that members will not be paid by OpenAI, and that the group can change its membership as it sees fit. The announcement states explicitly that the group will not be responsible for advising OpenAI on how to pace the company’s internal progress on mathematics.

OpenAI called working with the group a first step and said difficult questions remain about how AI can support mathematical understanding and how the benefits of these capabilities can reach the wider community, adding that it wants mathematicians to play the central role in shaping the answers.

The Group’s Charter and Current Task

On its own site, hosted at the Institute for Advanced Study, the group states that its purpose is to advise AI companies on their interactions with mathematical research and with the mathematical community — including the responsible presentation and release of mathematical results — and to act as one possible communication channel between mathematicians and the AI industry. The group writes that it believes the discipline is at a historic moment and that recent events raise urgent questions about how to support the long-term prospects for deep human understanding of mathematics.

The group states that it operates independently of any AI company, that members accept no payment for the work, and that it will publish its recommendations to AI companies on the site. It says it is willing to offer recommendations to any AI company whose models are likely to have a significant impact on mathematics, while noting that “we do not have decision making power at any AI company” and that responsibility for decisions rests with each company. According to the site, the group came together after OpenAI approached some of its members about establishing an external advisory board, and in agreement with OpenAI those members decided to create an independent group and invite others to join.

The initial members listed by OpenAI and by the group are François Charles (ENS-PSL), Camillo De Lellis (IAS, GSSI), Timothy Gowers (Collège de France, Cambridge), Martin Hairer (EPFL, Imperial College London), Nikhil Srivastava (Berkeley, Simons Institute), Ulrike Tillmann (Oxford, INI), Ravi Vakil (Stanford), Edward Witten (IAS), and Melanie Matchett Wood (Harvard). Hairer also appears among the signatories of the September 11 open letter.

The group’s stated current task is advising OpenAI on how to coordinate the release of what it describes as a large number of significant results in mathematics that OpenAI reports were produced by its internal model. The group is soliciting input from the mathematical community through a form on its site, and says responses will be used to inform its recommendations and will not be made public without approval.