**Tristan Buckmaster, an NYU mathematician, has publicly accused OpenAI of learning about his and Anthropic researcher Levent Alpöge’s year-long AI-assisted work on fluid dynamics equations (including progress toward Navier-Stokes Millennium Prize problems) and then rapidly launching a large-scale internal effort that produced a competing proof.** The dispute centers on timing, alleged pressure during calls with OpenAI researcher Sébastien Bubeck to exclude Alpöge from authorship due to his Anthropic role, and remarks interpreted as career threats, plus questions over whether private Codex sessions influenced OpenAI’s models despite disabled training settings. OpenAI denies direct access to user data or specific copying, acknowledges possible de-identified influence, and highlights its compute investment while recognizing the pair’s priority on related results. This fresh controversy over AI research credit, training data practices, and lab competition creates the main catalyst for any lawsuit speculation, with traders watching for further statements, formal complaints, or regulatory scrutiny in the coming days.