Terence Tao Reacts to OpenAI Math Drop
openai
| Source: HN | Original article
Terence Tao’s advisory group is scrutinizing OpenAI’s 722 newly released math manuscripts, including an unverified Navier‑Stokes claim that the Clay Institute has not confirmed.
OpenAI’s recent “math drop” – a batch of more than 700 AI‑generated manuscripts that claim solutions to a wide range of open problems, including a purported proof of the Navier‑Stokes millennium problem – has drawn a sharp rebuke from Fields Medalist Terence Tao.
Tao’s advisory group has begun a systematic review of the 722 submissions, noting that the Navier‑Stokes claim remains unverified by the Clay Mathematics Institute. In a brief public comment, Tao warned that “the indiscriminate use of AI is turning the subject into a meaningless production‑quota ‘game’ that ultimately …” undermines the deeper insight that traditionally drives mathematical discovery.
The reaction matters because OpenAI’s rollout, announced earlier this month alongside a separate release of AI‑generated solutions to over 370 longstanding challenges, signals a new scale of automated theorem‑proving. If such outputs are taken at face value, they could reshape research priorities, funding decisions and the very culture of proof verification. Yet Tao’s critique highlights a risk: AI‑crafted proofs may bypass the conceptual breakthroughs that give mathematics its explanatory power, making it harder for human experts to assess validity and relevance.
Going forward, the community will watch for the outcomes of Tao’s group’s scrutiny, any formal validation (or refutation) of the Navier‑Stokes manuscript by the Clay Institute, and OpenAI’s response to the criticism. Parallel commentary from AI‑skeptic Gary Marcus, also noting a “new regime in math,” suggests the debate will extend beyond academia into broader discussions about the role of generative AI in scientific research.
As we reported on Oct 8, OpenAI’s math model was already delivering large‑scale solution sets; Tao’s intervention now frames the next phase of scrutiny and policy dialogue.
Sources
Back to AIPULSEN