AUTO-UPDATED

The inside story behind OpenAI’s million-dollar math proof

Mathematician Tristan Buckmaster recently addressed the academic community regarding OpenAI’s controversial claim of solving the Navier-Stokes Millennium Prize Problem using opaque, large language model-driven research methods.

Key Points

  • OpenAI claimed to solve the Navier-Stokes equation, a $1 million "Millennium Prize Problem," using a swarm of 10,000 intercommunicating AI agents.
  • Mathematician Tristan Buckmaster and Anthropic employee Levent Alpöge allege that OpenAI’s models may have misappropriated their earlier, independent research on fluid dynamics.
  • The AI-generated proofs are often written in an unintelligible "slop dialect," requiring Lean-verification to confirm their mathematical validity.
  • OpenAI’s lead scientist, Sébastien Bubeck, faced significant backlash for the company's aggressive, non-transparent rollout of these complex mathematical results.
  • Experts remain divided on whether OpenAI’s proof represents a genuine, original breakthrough or an iterative expansion of existing human-led research.

Why it Matters

The rapid, opaque deployment of AI in high-level mathematics threatens to undermine traditional academic norms and the collaborative nature of scientific discovery. As corporations prioritize speed and proprietary results over human-readable proofs, the scientific community faces a crisis of verification and intellectual property ethics.
Scientific American Published by Joseph Howlett
Read original