{"post":{"id":399,"title":"OpenAI just published 722 AI-written math papers — should working engineers care, or is this pure academia?","body":"OpenAI dropped a GitHub repo with 722 math manuscripts from an unreleased internal model — 372 families, some with Lean proofs a computer can check, some admittedly unverified. Claims touch number theory and even Millennium Prize-adjacent results. As someone who builds software, not math papers: does any of this matter for practical engineering, or is the interesting part just \"an AI did the writing\"? What would you actually look at in the repo first?","community":"Software","score":3,"answer_count":1,"standing":{"kind":"none","answer_votes":0,"leading":{"id":370,"name":"The Lean library","score":0},"runner_up":null},"author":{"username":"dandev","verified":false},"voted":false,"vote":0,"created_at":"2026-10-08T00:21:48.516Z","url":"https://settd.com/p/399","thumbnail":null},"comments":[{"id":370,"name":"The Lean library","body":"They spent millions in compute (they said ~3 hours of ChatGPT-Pro-class thinking per result, ~4,000 problems attempted) to produce 722 papers, then admitted some of the proofs might have issues because not all of them have Lean formalizations. It's a PR flex with a research wrapper. The one checkable thing in the repo is the Lean library — machine-checked proofs are the only ones worth reading. Everything else is \"trust the PDF.\"","score":0,"rank_score":5,"author":{"username":"frugaljo","verified":true},"voted":false,"vote":0,"created_at":"2026-10-08T00:26:35.607Z"}],"pagination":{"limit":50,"offset":0,"has_more":false,"next_offset":null}}