Mathematician Steven Strogatz: 'I'm Really Terrified' by AI Math Breakthroughs
Original: ‘I’m Really Terrified’: A Mathematician Grapples With AI’s Recent Breakthroughs
Why This Matters
AI is now outpacing career mathematicians on landmark problems, reshaping who drives fundamental research.
Cornell mathematician Steven Strogatz spoke with WIRED in September 2026 about AI's seismic impact on mathematics, following OpenAI's claim of solving a 90-year-old $1M prize problem using tens of thousands of agents, and Anthropic's Claude formalizing 29,500 theorems related to Fermat's Last Theorem.
The past week in AI-driven mathematics has left Steven Strogatz—Cornell professor and coauthor of the forthcoming book Big Math—visibly emotional. OpenAI announced it used tens of thousands of AI agents to solve a 90-year-old problem carrying a $1 million prize, building on work by Spanish mathematicians Diego Córdoba and Luis Martínez-Zoroa. The announcement drew immediate controversy: mathematician Tristan Buckmaster alleges OpenAI rushed to publish after learning of his own parallel work with Anthropic researcher Levent Alpöge, and claims the company attempted to shape credit attribution. Days earlier, Anthropic reported Claude had proved 29,500 small theorems while formalizing an existing proof of Fermat's Last Theorem. OpenAI also announced progress on 10 other long-standing problems in August. Strogatz calls 2026 either an 'annus mirabilis or annus horribilis' for the field. His coauthor, Alex Townsend, has used ChatGPT to crack a decades-old numerical linear algebra problem—work he says would have been infeasible otherwise. Yet Townsend is blunt about the cost: 'I feel totally threatened by it. At a point in my career where I'm very productive and at my peak strength as a mathematician, that peak skill is no longer there.' Strogatz warns that future breakthrough mathematics will simply require AI. Corporate labs, he adds, are racing to capture headlines ahead of major IPOs—accelerating breakthroughs while straining the norms of academic credit and collaboration.