OpenAI announced on Tuesday that an internal AI model, described as significantly more capable than the current GPT-6 Astra, has developed a proof for the Navier–Stokes existence and smoothness problem. This mathematical challenge is one of the seven Millennium Prize problems, which carry a $1 million reward for a verified solution. The company claims the model successfully demonstrated that three-dimensional Navier–Stokes equations can develop a singularity in finite time, a finding that would represent a major milestone in fluid dynamics and computational science.

However, the announcement has been met with skepticism within the scientific community. Critics and independent researchers have raised questions regarding the provenance of the data used to reach this conclusion. Specifically, there are concerns that the AI may have relied upon or synthesized unpublished research from external mathematicians without proper attribution or authorization. This has sparked a broader debate about the ethics of using frontier AI labs' tools to solve complex scientific problems, particularly when the underlying data sources remain opaque.

While OpenAI frames this as a historic advancement in AI-assisted discovery, the controversy highlights a growing tension between the speed of AI development and the traditional standards of academic rigor. The situation underscores a core trust issue: whether the scientific community can verify findings produced by proprietary models when the training data and reasoning pathways are not fully transparent. As of now, the mathematical community is waiting for a formal peer review of the proof to determine if the AI's output holds up to rigorous scrutiny.