Terence Tao’s four-part essay asks what happens when AI can produce mathematical answers faster than people build understanding around them. A breakthrough traditionally starts conversations, collaborations, and clearer explanations. Tao argues that some AI-generated solutions are interrupting that process rather than accelerating it.

  • The person prompting an agent may not understand its proof well enough to explain it or answer other researchers’ questions.
  • Tao says fear of being scooped is encouraging researchers to keep promising directions private.
  • His proposed “Math 2.0” would reward explanation, community building, and opening new questions—not just being first with an answer.

This is not an argument against using AI in mathematics. Tao explicitly wants it involved in the work that turns a result into shared knowledge. His target is the incentive to harvest solutions and leave the understanding to someone else.

The 628-comment thread on Hacker News adds a historical parallel, a practical research constraint, and disagreement about where the technology goes next.

What the thread adds

  • sanxiyn — connects the argument to an older account of mathematical progress: “This is basically On proof and progress in mathematics by Thurston restated.” They link Thurston’s 1994 essay.
  • enum — brings the concern down to research planning: “The real question is how do you come up with a credible 3-5 year research program that is unlikely to be scooped or become pointless overnight.”
  • j-pb — proposes preserving access to the agents’ working context: “I want to "meet" the context that generated these proofs.”
  • pegasus — pushes back on that proposal: “When you’re asking them why they did this or that, they are basically guessing anew from the outside, and are just as likely to hallucinate as they are to hit the right answer.”
  • underdeserver — disagrees with the implied division of labor between machine answers and human explanation: “I just default to thinking the models are going to be much better than us at that, too.” That is a prediction, not a demonstrated result.

Handles are pseudonymous; HN publishes no per-comment scores. The source ordering is HN’s ranking, and this usefulness-selected slice of the thread is not a consensus.