How Fast Did AI Math Ability Grow Each Year?

Noam Brown says AI math ability grew about tenfold every year, measured by how long a human mathematician would need for the same kind of problem. Grade-school math, grades K through eight, takes a human mathematician about five seconds. The next year models did math benchmark problems that take an expert human mathematician about a minute. A year later they handled the qualifier for the USA Mathematics Olympiad team, about 10 minutes for a good mathematician. IMO Gold followed in 2025, and an IMO problem takes a human mathematician about 100 minutes, or an hour and a half.

On a trend line of 10x every year, an hour and a half at IMO gold becomes 15 hours the next year, still not enough to solve a Millennium Prize problem. He did not expect a solution in 2026, probably not in 2027, and only maybe in 2028. It happened a lot faster than he expected.

The claim that models are replacing mathematicians, or that they are superhuman in mathematics across the board, is the wrong takeaway. They are exceptional in some ways and weaker than human mathematicians in others: brilliant in some dimensions and weaker in others. They are not very good at posing new problems or at judging which branches of mathematics are worth exploring. He would be thrilled if AI stays a complement to human abilities and helps discover new knowledge without fully replacing people.