UC Berkeley math community left at crossroads in wake of OpenAI Navier-Stokes proof

Facebook Twitter WhatsApp LinkedIn SMS Email UC Berkeley math professor and Fields medalist Richard Borcherds said he only expected capabilities of this caliber in 10 years. Now, mathematicians are torn on how AI will reshape their field. Anjalie Butte | Staff Facebook Twitter WhatsApp LinkedIn SMS Email Print Copy article link Save Facebook Twitter WhatsApp LinkedIn SMS Email Print Copy article link Save After OpenAI announced its solution to one of mathematics’ most difficult unsolved problems, the Navier-Stokes existence and smoothness problem, UC Berkeley mathematicians are considering AI’s impact on the future of the field. Announced in 2000, the Navier-Stokes problem is one of seven “Millennium Prize Problems” — unsolved questions chosen by the Clay Mathematics Institute to celebrate “mathematics in the new millennium.” Anyone who solves one of the problems is awarded $1 million. OpenAI’s solution, produced by an internal AI model, used approximately 10,000 AI agents, or autonomous AI systems. OpenAI research scientist Noam Brown said in a post on X that the result cost “millions of dollars,” exceeding the prize amount itself. Yes, this result cost millions of dollars. But remember that when @OpenAI announced o3 it cost ~$500,000 to score 87.5% on ARC-AGI 1. Today, Astra scores higher for ~$20. In 2025 it took us and GDM an enormous amount of compute to achieve IMO gold. For the 2026 IMO, anyone with a $20/month ChatGPT subscription could do it. Massively scaling test-time compute gives us a glimpse of the future. I believe that a year from now everyone will have an AI at their fingertips capable of solving problems of this caliber. — Noam Brown (@polynoamial) September 8, 2026 UC Berkeley math professor Richard Borcherds, who won the 1998 Fields Medal, a prize considered to be the math equivalent to a Nobel, said he only expected a result of this caliber in 10 years and is surprised at the pace at which AI is improving. “It’s fairly obvious that if things go like this, in a few years, humans doing math research will be redundant,” Borcherds said. It’s hard to know what advice to give graduate students because the future is “changing in a rather major way,” he added. Borcherds claimed AI is already better than beginner graduate students. “If you’re starting off as a (graduate) student in math, it’s likely that AI will always be better than you at mathematics,” Borcherds said. “Grad students are not stupid — this has occurred to all of them, and quite a lot of them are dropping out or getting very worried about it.” However, Robin Huang, who is in the final year of her math Ph.D. program at UC Berkeley, is not convinced of AI’s capabilities just yet. Though she has used AI to aid her own research, she has found it “very stupid and very inefficient.” Huang said she thinks many graduate students and postdocs in math are concerned about job prospects, but she remains optimistic about the future. In fact, in a “perfect vacuum,” she said, if AI’s abilities were to stay the same, she would be able to learn a lot more about mathematics. Huang said AI companies’ pursuit of unsolved math problems is an “extremely cynical undertaking” and said she saw AI having a negative impact on the math community. She accused AI companies of turning mathematicians against each other and creating fear about being “scooped,” where another researcher — or, more recently, an AI model — publishes results before a rival does. OpenAI’s Navier-Stokes solution is currently drawing scrutiny from mathematicians, notably New York University math professor Tristan Buckmaster, who posted a statement on social media alleging that the AI models may have been trained on Buckmaster and his collaborators’ prompts in OpenAI’s Codex. He also alleged OpenAI refused to credit mathematician Levent Alpöge due to his ties to rival AI company Anthropic. Borcherds remains unfazed by AI and said these advancements make no difference if one enjoys learning math. “People are complaining that AI is pushing out this 100,000-line formal proof of some major conjecture. I don’t really see what harm this does because you can always ignore this if you want, and then you’re no worse off than before,” Borcherds said. “I think what people are getting upset about is they’re slowly realizing that they are dumb and somehow blaming this on AI, even though it’s not really AI’s fault.” Borcherds, who currently teaches MATH 250A, a graduate-level course on abstract algebra, also said AI would be much better than him at teaching the course. He said AI is more patient and much better at expressing itself than he is and doesn’t get mad when students ask stupid questions. Borcherds himself said he has asked AI “deep mathematical questions” and gets “much better answers from it than (he gets) from any human mathematician.” Huang tells her students that math is half about communicating with others — a point Borcherds echoed, as he said nearly everything he has learned about mathematics was taught to him by others. But Borcherds added that it doesn’t “(make) much difference” whether he learns math from AI or from other mathematicians. Huang said she believes there is something “deeply creative” about math, which she argues isn’t true for the lab sciences. Math is about coming up with one’s own perspective of the world, which Huang said AI can’t do. “I’ve told my students that you can, by all means, use AI,” Huang said. “I don’t care if you do your homework or not. But if you don’t do your homework by yourself, you will fail the f--- out of the exam.” Huang mentioned discussions about the math community’s career incentive scheme being flawed and mathematicians falling victim to the idea that math is about proving big unsolved problems. “Maybe AI companies are able to do this because our community standards and incentives were already wrong,” Huang said. “We already knew this, and maybe this will be a wake-up call for us. I’m not saying that I’m confident that everything will change immediately, but I do have faith that we will be able to move past this. … The other thing is, there are other things in life that make me happy. Even if — which I do not think will happen — but, even if (AI) completely takes over math, there are things in life that may not be, and life goes on.” Facebook Twitter WhatsApp LinkedIn SMS Email Print Copy article link Save Tags Openai Navier-stokes Millennium Prize Problems Clay Mathematics Institute X Noam Brown Ai Artificial Intelligence Uc Berkeley Fields Medal Richard Borcherds Robin Huang Tristan Buckmaster Anthropic Codex Levent Alpöge Litong Deng Deputy Night Editor Author linkedin Author twitter Author email Follow Litong Deng Support Our Nonprofit Newsroom We're an independent student-run newspaper, and need your support to maintain our coverage. Become a member today! Support! Sign up for our email newsletter! * indicates required Email Address * First Name * Last Name * Newsletter Subscription Daily Newsletter Weekly Newsletter Local Events Latest Strikeout Punctuating life with depression 4 hrs ago Literature Her’s biographer Patrick Breen on the Liverpool indie-pop duo’s timelessness 4 hrs ago Film & Television ‘You Can See Everything’ review: on the Holmes front 5 hrs ago Film & Television ‘Capturing Bigfoot’ team talks everything but Bigfoot 5 hrs ago The Soapbox Climate crisis or emergency? The difference matters 6 hrs ago