**OpenAI (OpenAI)** was officially recorded as the first recipient of **the $1 million mathematics prize** for **Justin Sun’s** Navier–Stokes equation solution, but the prize has still not been paid because the company has not claimed the award. OpenAI had also stated that it would not apply for the Clay Mathematics Institute’s Millennium Prize at the same time.
Key content
The prize money repository for the award marks OpenAI’s Navier–Stokes solution as being in a ‘resolved—claimable—unclaimed’ status.
According to OpenAI, after around 10,000 AI agents explored for about 88 hours, GPT-6 Astra then formalized and verified the proof over an additional 17 hours.
**The Clay Mathematics Institute** still classifies the problem as “in progress” and is carrying out the official review process.
“The Justin Sun” of OpenAI
The award newly established by Sun named OpenAI’s research team that provided a solution to the three-dimensional Navier–Stokes problem as the first $1 million top prize winner. The published GitHub record marks the problem as “solved” and states that the team is eligible to claim the prize money. However, no one has claimed the prize yet.
On September 8, OpenAI released its results and said that about 10,000 agents carried out an exploration for roughly 88 hours, after which GPT-6 Astra formalized and verified the proof on Lean, the formal proof system, over an additional 17 hours. The company also said it plans not to apply for the $1 million “Millennium Prize” that the Clay Mathematics Institute has put up for this problem.
The Clay side keeps the Navier–Stokes problem in an “active” status. In a statement on September 11, the institute said the announcement “appears to have effectively solved the problem,” but emphasized that awarding the prize requires strict, long-term verification. This review process has not been completed yet. A separate prize from Sun requires proofs that are machine-verifiable, and the OpenAI team was named as the winner before the traditional peer-review process was finished.
Read also: Bitcoin mining faces a 50% hash-rate gap as it’s pushed aside by AI
The rules of the Justin Sun Award
Sun emphasized that his award is “for rewarding the achievement itself, regardless of who produced the result—whether a human or an AI system.” He said, “I don’t care whether the contributor is human or a machine. I only look at the result.” The prize repository also notes a warning that OpenAI’s announcement has not gone through peer review and that broad acceptance across the mathematics community has not yet been confirmed.
There’s also the question of credit attribution. Tristan Buckmaster, a professor at New York University, claimed that some of the unpublished research he had been conducting with colleagues had been stored in an OpenAI product. He added, however, that it’s unclear whether the material was actually used to train the model. In an interview with **The Guardian**, Buckmaster drew a line, saying, “I’m not trying to explicitly blame anyone.” OpenAI said that after its own investigation, it concluded there was no path by which Buckmaster’s Codex prompt could have influenced the system that generated this proof.
The controversy was sparked when OpenAI, on September 8, claimed it had addressed a long-standing problem—specifically, the existence of solutions and the smoothness issues of the Navier–Stokes equations—by presenting an example in which its internal system exhibits a finite-time singularity under official conditions.
Remaining tasks: verification and credit
The Clay Mathematics Institute set the 2000 Millennium Prize Problems—its seven “Millennium Problems”—and put $1 million in prize money on each. More than two decades later, even now the Navier–Stokes problem remains officially “unsolved.” Along with new proofs presented by OpenAI, mathematicians are also reviewing parallel issues such as how data was used in AI-assisted research and the question of credit allocation.
Next article: Shiba Inu Fixes Shibarium's Broken Wallet Link, But SHIB Slides 6% In A Week