A year after questioning LLM conjecture-solving, Cornell professor concedes AI surged ahead

burny_tech · x · 2026-09-20

Cornell professor Amin Karbasi and a theoretical CS colleague took on a state-of-the-art AI theorem prover. A year ago his paper questioned whether LLMs could solve easy conjectures — a take that 'did not age well.' 'We are already far beyond the world in which we asked that question. The future arrived earlier than we thought.'

Related event: Professor Admits Paper Doubting LLM Math Skills Aged Poorly(2 posts)→

Original post →

More from Research

Research channel →