How many more Millennium Prize Problems will AI solve in 2026?
I think there is a meaningful chance that no eligible AI lab will make a qualifying year-end claim on one of the five listed Millennium Prize Problems, but the recent OpenAI activity has materially lowered that certainty. My estimate is that the market should lean toward 0 more being the outcome, though not overwhelmingly so.
Analysis
The central fact is that the market only resolves Yes if an eligible AI lab publicly announces that it has solved one of the five qualifying Millennium Prize Problems by the end of 2026. The recent OpenAI Navier-Stokes claim does not count for this market, so the key question is whether a qualifying announcement for Birch and Swinnerton-Dyer, Hodge, P versus NP, the Riemann Hypothesis, or Yang-Mills will occur in the remaining time. Based on the current information, there is real momentum in AI-assisted mathematics, but the gap between a high-profile claim and a qualifying, properly scoped announcement on one of these specific problems is still substantial. That makes zero additional qualifying announcements the more likely outcome, even though it is far from guaranteed.
Historically, Millennium Prize Problems are extraordinarily resistant to quick resolution, and the remaining candidates are all deeply entrenched problems with large, skeptical expert communities. The recent OpenAI report suggests that modern models can produce striking mathematical outputs and may be approaching frontier-level reasoning, but the market only pays out on an explicit solved announcement from an eligible lab, not on substantial progress, partial derivations, or outside validation. That distinction matters a lot here, because labs tend to be cautious with the strongest claims when the proof burden is enormous and reputational risk is high. Even if one of the problems is genuinely cracked internally, turning that into a public announcement within the calendar year requires enough confidence, verification, and editorial discipline to survive scrutiny, which is a much higher bar than producing impressive internal results.
The market price is currently implying that a non-zero outcome is somewhat more likely than zero, which may reflect the shock value of the OpenAI math news and the possibility that another lab could follow with a year-end breakthrough. I think that optimism is understandable, but still somewhat aggressive. The most plausible path to a qualifying Yes is a rapid cascade from the current math breakthrough into a second, clearly qualified announcement on one of the eligible problems, most likely Hodge or possibly Riemann Hypothesis if a lab already has a serious internal claim. The more likely path, however, is that the year ends with one or more major claims still under review, or with progress being touted without the exact language needed to qualify. That leaves 0 more as the stronger outcome overall.
Arguments
For
- Arguments for Yes: Recent AI mathematics progress suggests the probability of a breakthrough announcement is no longer negligible.
- Arguments for Yes: The market’s current pricing indicates traders see a meaningful chance that one qualifying claim lands before year-end.
Against
- Arguments against Yes: The current headline claim is about Navier-Stokes, which is explicitly excluded and therefore does not count.
- Arguments against Yes: The remaining listed problems are extremely hard, and there is little evidence of a completed, publicly announced solution to any of them.
Key drivers
- The qualifying bar is strict, because the lab must publicly announce that a listed Millennium Prize Problem has been solved.
- The recent OpenAI claim shows progress is possible, but Navier-Stokes is excluded and does not itself affect this market.
- There is limited time left in 2026 for a second lab to produce, verify, and announce a qualifying solution.
- Millennium Prize Problems are exceptionally difficult, so even major AI math advances do not translate easily into confirmed resolutions.
Risk factors
- A lab could unexpectedly announce a qualifying solution to Hodge, Riemann Hypothesis, or another listed problem before year-end.
- The current wave of AI math optimism could encourage multiple labs to push out stronger claims than expected.
- If OpenAI or another eligible lab already has a near-finished proof, the remaining verification and announcement step could happen quickly.
Scenarios
Best case
An eligible AI lab quickly finalizes and announces a solution to one of the five qualifying problems, most plausibly Hodge or Riemann Hypothesis, before the end of 2026, making the answer 1 or more rather than 0.
Most likely
AI continues to produce major math-related claims and partial results, but no eligible lab makes a clean, qualifying solved announcement on one of the five listed Millennium Prize Problems before year-end, so the market resolves to 0 more.
Worst case
The current math excitement fades into continued internal work, with all public statements remaining at the level of progress, claims under review, or excluded problems like Navier-Stokes, leaving the final count at 0.
More from this day
- PoliticsKalshi20d
Will 3 mg THC beverages qualify as hemp under federal law?
AI9%MKT87%Edge-78HypedA 3 mg THC beverage is very unlikely to qualify as hemp under the federal changes taking effect in November 2026. My independent probability for a Yes outcome is 9%, far below the market price.
- PoliticsKalshi2mo
When will a catastrophic AI duty of care become U.S. law?
AI58%MKT8%Edge+50Hidden GemI put the odds of Yes meaningfully above the market because several state laws now create AI safety obligations that plausibly fit a catastrophic-risk duty of care before the deadline. The main reason not to go much higher is that the exact contract wording may be read narrowly as federal law only, which would leave the outcome much less likely.
- PoliticsKalshi2mo
When will the debt limit be increased?
AI68%MKT24%Edge+44Hidden GemI think the market is substantially underpricing this event. Based on current reporting and the usual congressional pattern, there is a better-than-even chance that some form of debt-limit increase becomes law before Jan. 1, 2027.