OpenAI stole mathematicians' private research from their own Codex chats
nreece
18 points
4 comments
September 08, 2026
Related Discussions
Found 5 related stories in 128.5ms across 10,409 title embeddings via pgvector HNSW
- OpenAI fought dirty on career-making math problem jonbaer · 161 pts · September 08, 2026 · 68% similar
- OpenAI might have stolen another major proof tamnd · 59 pts · September 10, 2026 · 68% similar
- Top mathematicians are outraged by OpenAI's methods sbulaev · 91 pts · September 11, 2026 · 67% similar
- More questions about whether researchers can trust OpenAI with unpublished math pred_ · 769 pts · September 10, 2026 · 66% similar
- Google copied our open-source code, removed engineers' names without credit jacquesm · 123 pts · September 15, 2026 · 65% similar
Discussion Highlights (4 comments)
paxys
Keep this kind of clickbait on Reddit please
peesem
regardless of the validity of the claim, why post a link to Reddit when the Reddit post is just a link to a document? why not directly link to the document
lioeters
The actual link: https://cims.nyu.edu/~tristanb/statement.pdf
ihuman
Dupe: https://news.ycombinator.com/item?id=49605915