Fields Medalists warn of AI risks to mathematics
In a joint warning letter, 25 prominent recipients of the Fields Medal, widely regarded as the highest distinction in mathematics, cautioned that the competition among AI companies to solve renowned mathematical problems could have damaging long-term consequences for the discipline, according to The Economist.
The letter, signed by leading mathematicians including Terence Tao, came days after OpenAI announced that one of its AI systems had solved one of the Millennium Prize Problems, a set of mathematical challenges that have remained unresolved for generations.
“No alignment” between AI companies and mathematics
The strongly worded letter stated that the objectives of AI companies and those of the mathematical community are “largely incompatible,” warning that this trend poses “a general threat to intellectual work.”
The 25 signatories represent nearly five decades of the history of the Fields Medal, from Pierre Deligne, who received the award in 1978, to Yiming Deng, who was honored this year.
The mathematicians said mathematical problems are part of a broader system in which challenges are posed to help students develop their skills. “Solving problems is merely a tool and a means toward the ultimate goal of theoretical understanding,” they said, emphasizing that these processes “always take time and are based on human interaction.”
The letter raises the question of “how to ensure that, as AI changes the way work is done, we do not lose sight of what that work was intended to achieve.” It stressed the need to address these issues urgently within the mathematical community, among companies developing such technologies, and across society more broadly.
The warning follows OpenAI’s announcement last Tuesday that one of its unreleased models had successfully solved the Navier–Stokes problem, one of the seven Millennium Prize Problems. The process reportedly took 88 hours and involved thousands of AI agents working in parallel.
