AI solves 80-year-old math problem, reshaping research forever

An artificial intelligence has solved a mathematical problem that stumped researchers for 80 years, sending shockwaves through the global mathematics community. The breakthrough, achieved by an AI system developed by an unnamed team, marks the first time machine learning has cracked a long-standing open question in pure mathematics, according to reports in *Svenska Dagbladet* and *Dagens Nyheter*.

Svante Linusson, professor of mathematics at Sweden’s Royal Institute of Technology (KTH), called the development "still a bit of a shock." "We have spent decades trying to solve this, and now an AI does it in what seems like a matter of weeks," he told *Dagens Nyheter*. The problem itself remains unnamed in the reports, but its resolution is described as a milestone in the intersection of mathematics and artificial intelligence.

Experts warn the achievement could accelerate a broader shift in how mathematical research is conducted. Linusson noted that while the AI’s methods are not yet fully understood, the speed of progress raises urgent questions about the future role of human mathematicians. "We have many crises in the world right now," he said. "But when we look back on our time, it may be AI that we should have focused on most."

The breakthrough comes amid growing debate over AI’s expanding capabilities in fields traditionally dominated by human intellect. Earlier this year, AI systems demonstrated proficiency in generating original proofs and identifying patterns in complex datasets, but this is the first confirmed case of an AI solving a problem that had resisted human efforts for nearly a century.

No details have been released about the AI’s architecture or the specific mathematical field involved, though speculation centers on areas like number theory, combinatorics, or algebraic geometry—domains where long-standing conjectures have remained unproven. The development follows a 2023 announcement by Google DeepMind, which used AI to solve a 50-year-old problem in knot theory, but the 80-year timespan of this latest achievement underscores its historic significance.

Mathematicians are now grappling with the implications. Some see AI as a powerful tool to augment human intuition, while others fear it could render certain types of research obsolete. "This isn’t just about solving one problem," said a researcher quoted in *Svenska Dagbladet*. "It’s about whether AI can redefine what it means to *do* mathematics."

The identity of the AI’s creators and the problem’s exact nature are expected to be disclosed in a forthcoming peer-reviewed paper, though no publication date has been set. For now, the mathematics community remains in a state of cautious excitement—and unease—over what may be the first of many such breakthroughs.