Traders now expect OpenAI to announce another Millennium Prize solution before 2028

Traders have shifted from a near-even view to favoring an OpenAI announcement of another Millennium Prize solution before 2028. Bets rose further Tuesday after climbing on Monday, but no verified news explains the latest jump. The shorter-term market, which asks about an announcement this year, still treats one as unlikely.

Expectations dropped sharply on Saturday, briefly tilting against an announcement, then rebounded later that day. They eased again on Sunday before turning higher on Monday. Tuesday’s rise carried the outlook from close to even to likely. The swings follow earlier sharp drops and recoveries that also had no confirmed explanation.

OpenAI’s recent math announcements provide context but have not named another eligible prize solution. On Sept. 8, the company said an internal model had produced a proof related to the Navier-Stokes problem. That problem is excluded from this question, and OpenAI said it was not claiming the Clay Mathematics Institute prize for the work. On Sept. 21, OpenAI said the model had resolved more than 100 other open problems and announced an independent mathematics advisory group, without identifying another qualifying solution. The claims also drew scrutiny over the work’s significance and origins, according to Nature and The Washington Post.

A qualifying announcement would need to say OpenAI, its researchers or its models helped solve one of five problems: the Riemann Hypothesis, P versus NP, Yang-Mills existence and mass gap, the Hodge Conjecture, or the Birch and Swinnerton-Dyer Conjecture. Partial progress would not count. OpenAI has no scheduled announcement or other direct event tied to the question; further math results could shift expectations if they identify a solution to one of those problems.