Why I won’t leave mathematics, despite AI solving centuries-old problems | Ken Ono

The Guardian
Read full story· 5 min
Ken Ono reflects on nearly quitting mathematics after AI-assisted proof of the Riemann hypothesis, but finds renewed purpose using AI to formalize and verify complex proofs, blending human insight with machine precision.

What happened

Ken Ono, a mathematician, describes his experience after an AI-assisted proof of the Riemann hypothesis was announced in November 2023 by Levent Alpöge using Claude. The proof was formally verified by the Lean and Mathlib communities. Ono, who considered leaving mathematics, instead began working with AI to formalize and verify complex mathematical proofs at Axiom Math. He and his team released the first complete Lean formalization of the classification of finite simple groups, using AI to produce over 43 million lines of code and uncovering errors in previous proofs.

Key details

  • AI-assisted proof of the Riemann hypothesis announced on November 20, 2023, by Levent Alpöge with Claude.
  • Lean and Mathlib communities certified the formal proof.
  • Ken Ono is on leave from the University of Virginia and works at Axiom Math.
  • The Lean formalization of the classification of finite simple groups required 43.1 million lines of code.
  • A problem was found and corrected in a previously published proof during the formalization process.

Why it matters

Mathematicians, students, and academic institutions now face new ways of working with AI in research, including formalizing and verifying proofs at a scale not previously possible. This changes how mathematical contributions are made, credited, and checked, and introduces new standards for transparency and accuracy. Those entering mathematics must navigate a landscape where AI tools are integral, and institutions must decide how to support and evaluate work that involves or excludes AI.


More in AI Research