Will Artificial Intelligence solve a Millennium Prize Problem before 2030?
💡 What the odds say
The market puts this at about a 59% chance — more likely than not.
No money — just record your call and see if you were right. Yes is at 59% right now.
The market is pricing a surprisingly high chance that AI will crack one of the seven Millennium Prize Problems by 2030, driven by a string of recent AI breakthroughs on long-standing math problems, but the leap from solving isolated hard problems to a Millennium Prize problem remains immense and historically unprecedented.
📊 Base rate: No AI system has ever solved a Millennium Prize Problem, and the seven problems have stood for decades (some over a century) despite intense human effort, giving a historical prior well below 10% for any single problem being solved by any method within a 4-year window.
What's driving it
- • OpenAI's breakthrough on an 80-year-old math problem (The Guardian, May 21) directly shows AI can tackle problems that resisted human mathematicians for decades, raising expectations for even harder targets.
- • A Forbes report that AI solved a problem that had stumped the world's best minds for decades (Forbes, Apr 17) adds to a pattern of AI surprising experts in pure math, shifting the perceived difficulty curve.
- • The Conversation's piece on a 'golden age' of mathematics thanks to AI (Jul 23) and IEEE Spectrum's exploration of what it means to be a mathematician when AI does the math (Jun 25) reflect a growing narrative that AI is fundamentally changing mathematical discovery, which may be inflating odds beyond concrete evidence.
- • No clear catalyst for the current 59% level beyond the cumulative effect of these headlines; the odds have likely risen gradually as each new success story is published.
The case for YES
- • AI has already solved multiple long-standing math problems in 2026 (e.g., the 80-year-old problem OpenAI solved, per The Guardian, May 21), demonstrating a capability that could scale to a Millennium Prize Problem if the right approach is found.
- • The rapid pace of AI progress in mathematics, with systems now able to simulate nature faster (Communications of the ACM, Jun 1) and generate novel proofs, suggests that a breakthrough on a Millennium Prize Problem is within reach before 2030 if current trends continue.
- • The 'golden age' narrative (The Conversation, Jul 23) implies that human-AI collaboration is unlocking new methods, and a Millennium Prize Problem could be the next target as researchers actively apply AI to these hardest problems.
The case for NO
- • The seven Millennium Prize Problems are qualitatively harder than any problem AI has solved so far; they require fundamentally new concepts or frameworks, not just clever computation or pattern recognition, as highlighted by the IEEE Spectrum piece (Jun 25) on the changing role of mathematicians.
- • The experiment of spending a year asking AI to solve math's hardest problem (Communications of the ACM, May 29) likely ended in failure or limited progress, suggesting that current AI still falls short on problems of this caliber.
- • Even if AI makes incremental progress, the 2030 deadline is only 4 years away, and no AI system has yet demonstrated the ability to produce a complete, rigorous solution to a problem that has resisted the world's best minds for decades or centuries.
What to watch
- • Any announcement from a major AI lab (OpenAI, DeepMind, etc.) of a new technique specifically targeting a Millennium Prize Problem would push odds sharply higher (direction: Yes).
- • A peer-reviewed publication showing AI solved a different long-standing but non-Millennium problem would moderately increase odds (direction: Yes).
- • A prominent mathematician publicly stating that current AI approaches are fundamentally insufficient for Millennium Prize Problems, or a major conference concluding no progress, would push odds lower (direction: No).
AI-generated · grounded in recent news + odds · informational only, not advice. Verify on the source platform.
Data from Manifold’s public API, for informational purposes only. PredictPal is not affiliated with any platform and does not facilitate trading.
Discussion
Loading…
How it resolves
Resolved by whoever created the market, at their discretion per the question's description. It's play-money (Mana) and not tied to an official source — treat it as a community forecast.
ⓘ A market settles under its own written rules, which can lag what looks decided in the news — so the price may not move to 100% the moment an outcome seems obvious.
View the official rules on Manifold ↗Related markets
BNB Up or Down - Daily
The market's 70% odds for BNB rising today are not obviously supported by any of the provided recent headlines, which predominantly cover Airbnb regulatory news and a single crypto-related price drop from five months ago.
Yes ≈ 81% chance
DOGE Up or Down - Daily
Despite a cascade of bearish headlines—including a 25% weekly plunge in early June and a down close on July 11—the market now assigns an 80% probability to DOGE closing up on the daily resolution, suggesting traders are betting on a sharp reversal or a catalyst not captured in the provided news.
Yes ≈ 91% chance
Polymarket or Kalshi?
This market is a pure sentiment gauge with no resolution, so the 72% Polymarket lead reflects a persistent narrative advantage from its on-chain identity and aggressive bonus campaigns, but Kalshi's recent $500 referral push and a Washington Post controversy show it can still capture attention and shift the conversation.
2 outcomes
Will over 100,000 people be conceived with the help of advanced embryo selection techniques by 2030?
The market sees advanced embryo selection as still niche and ethically fraught, but recent coverage of genetic optimization startups and IVF's expanding role in family planning suggests the technology is gaining momentum, making the 100,000-conception threshold plausible if regulatory and adoption barriers fall.
Yes ≈ 26% chance