Quote · Hard Fork
Hot I.P.O Summer + What Is A.I. Doing to Math? + HatGPT
Where this was said
The Unit Distance Conjecture: AI's First Truly World-Class Math Result
At 39:18 · chapter starts 36:00
The discussion of Erdős benchmarks builds to its climax: OpenAI's resolution of the unit distance conjecture. Hartnett explains what made this result different from the string of solved Erdős problems that preceded it. Other solved problems had been largely ignored by serious mathematicians — 'sophisticated riddles,' in Hartnett's framing — but the unit distance conjecture was one that many mathematicians had actively tried to solve, failed, and moved on from [1] — Kevin Hartnett "OpenAI's proof of the unit distance conjecture wasn't just another solved Erdős problem. Mathematicians had actually tried to crack it; the…" 38:20 . The methods OpenAI's model used were sophisticated, non-obvious, and genuinely surprising. Most importantly, the mathematical community agreed nearly unanimously that the result was publishable in the Annals of Mathematics, the top journal in the field. Hartnett describes the pattern that had developed before this proof: 'AI did this, but it can't do that. Oh, it did that? But still can't do this.' The unit distance conjecture proof ended that goalpost-shifting. AI, Hartnett concludes, 'can do absolutely top-tier research.'
OpenAI's proof of the unit distance conjecture wasn't just another solved Erdős problem. Mathematicians had actually tried to crack it; the methods were sophisticated and surprising; and the result was universally agreed to be Annals of Mathematics quality — the top journal in the field. This ended the goalposts-moving: AI can do world-class research.
At the Institute for Advanced Study, Kevin Hartnett met two top mathematicians in one afternoon with polar opposite views — one dismissed AI entirely, the other said it would make mathematicians obsolete in 2 years. Terence Tao, the most important figure in the field, sits in the middle: AI as an Iron Man suit for mathematical thought.
