‘World’s greatest living mathematician’ shares his thoughts on AI

Terence Tao, a University of California, Los Angeles, professor often dubbed the “world’s greatest living mathematician,” took to microblogging site Mastodon last month to share his insights on the capabilities of OpenAI’s new GPT-o1 reasoning model, suggesting that while impressive, it still falls short of replacing human mathematicians. The “Mozart of Math,” renowned for his Fields Medal-winning work and contributions to various mathematical fields, conducted experiments and found that while GPT-o1 demonstrates improved reasoning abilities compared to its predecessors, it still struggles with advanced mathematical tasks and requires significant guidance.

Not there yet: In his series of posts, Tao likened the experience of working with GPT-o1 to “advising a mediocre, but not completely incompetent, graduate student.” He noted that the model could successfully navigate complex problems with “a lot of hints and prodding” but failed to generate key conceptual ideas independently. Tao shared in an interview with The Atlantic that human graduate students possess the capacity to learn and grow, while AI “sometimes just snaps back to the thing it tried before.”
Promising future: Despite these limitations, Tao remains optimistic about the future of AI in mathematics. He envisions a collaborative relationship where AI tools assist mathematicians in exploring large-scale problems and automating tedious tasks. “You might have a project and ask, ‘What if I try this approach?’ And instead of spending hours and hours actually trying to make it work, you guide a GPT to do it for you,” Tao explained.

Share this Article