OpenAI Claims Navier-Stokes Math Breakthrough Amid Plagiarism Accusations

OpenAI announced that its artificial intelligence agents solved the Navier-Stokes existence and smoothness problem, which is one of seven Millennium Prize Problems selected in 2000. [1] The proof demonstrates that the Navier-Stokes equations can occasionally "blow up," predicting an impossible state where a fluid's speed becomes infinite. [1][2] The company stated that the proof was verified using the programming language Lean, which practically guarantees its correctness. [2] Prior to this announcement, only one other Millennium Prize Problem had been solved. [1]

The achievement has been overshadowed by accusations from mathematician Tristan Buckmaster, who alleged that OpenAI utilized uncredited, AI-assisted work completed by him and Levent Alpöge. [1] Buckmaster posted that he and Alpöge had solved a related step involving the Euler equations, and he claimed OpenAI adopted their method after learning of their progress. [2] OpenAI denied these allegations, with staff member Sébastien Bubeck stating the company's model solved the Euler problem independently through different means. [2] However, Bubeck also noted that OpenAI was inspired to pursue the problem after hearing rumors about the efforts of Buckmaster and Alpöge. [1]

This development suggests that frontier AI models are becoming essential tools for solving complex mathematical problems, potentially challenging traditional norms of academic collaboration. OpenAI's success follows previous high-profile AI math achievements but sets a new benchmark by claiming a Millennium Prize problem and verifying it through the Lean programming language. Unlike past solutions developed slowly by individuals within the academic community, this breakthrough indicates that elite AI companies now wield the proprietary resources to outpace traditional institutions. What the sources don't address: whether OpenAI intends to formally claim the one million dollar prize associated with solving the problem.