Navier-Stokes “Solved by 2027” Bet Surges to 79% on Manifold After a Viral Claude Rumor. A prediction market asking whether the Navier-Stokes existence and smoothness problem will be considered solved on June 3, 2027 has jumped to a 79% implied chance of “Yes” as of September 08, 2026, with roughly half of its all-time volume trading in a single day and 99 unique bettors active. The spike lines up with a viral social-media claim that Anthropic’s Claude cracked the century-old fluid-dynamics problem, a claim that no institution has confirmed and that one of the world’s leading mathematicians has publicly walked back.
What the market is asking
The contract in question lives on Manifold, a play-money forecasting platform where users trade on the probability of future events. The question is blunt: “Will Navier-Stokes be considered solved on June 3th 2027?” To resolve “Yes,” the community needs to agree that the Navier-Stokes existence and smoothness problem, one of the six unsolved Clay Millennium Prize Problems, has an accepted proof by that date.
Navier-Stokes describes how fluids move. The open question is whether smooth starting conditions always stay smooth, or whether solutions can blow up into a singularity in finite time. It has resisted proof for decades, and the Clay Mathematics Institute still lists it as open with the $1 million prize unclaimed. If you are new to how these contracts price uncertainty, our explainer on how prediction markets work covers the mechanics.
The odds, as of September 08, 2026
As of September 08, 2026, the market showed a 79% implied probability of “Yes” on Manifold. The reason it is worth writing about is not the level alone but the flow: about 50% of the contract’s all-time volume traded in the prior 24 hours, one of the highest single-day volumes in the source data, spread across 99 unique bettors. That is the signature of a market reacting to news rather than drifting.
A word of caution on reading the number. Manifold runs on play money, so its prices reflect attention and conviction more than dollars at risk. An implied probability is simply the market price read as a percentage, a concept we unpack in implied probability explained. A 79% reading says the crowd currently leans heavily toward “Yes,” but it is a crowd trading on a rumor, not on a published proof.
What is actually driving the move
The catalyst is traceable to a single post. On September 5, 2026, AI commentator Andrew Curran wrote on X: “It’s fun to make predictions. Here’s a new one: Anthropic has solved a Millennium Prize Problem. And I’ll be even more specific. Claude has solved Navier-Stokes. It is out for expert review. And to give myself a hard deadline, they will announce it before the IPO.” The post was explicitly framed as his own prediction, cited no source, and used a self-imposed deadline as a scoring mechanism.
The message spread quickly, with trending pages tracking millions of views, and it collided with a real backdrop: Anthropic has produced genuine, verified mathematical results this year. That combination, a confident forecaster plus a company with a live track record in AI-assisted math, is what pushed speculative interest, and the Manifold price, sharply higher.
Who and what is involved
Several named figures anchor the story. Andrew Curran is the forecaster whose post started it. Terence Tao, a Fields Medalist and one of the leading analysts working on fluid equations, gave the rumor early lift when he discussed a hypothetical scenario in which an AI might solve a landmark problem using hidden computation. Tao later clarified that the scenario was hypothetical, that he is not aware of any confirmed breakthrough on Navier-Stokes, and that he never said Claude solved it.
On the corporate side sits Anthropic, the maker of Claude, whose expected initial public offering is the “hard deadline” in Curran’s post. Coverage has also referenced mathematician Tristan Buckmaster in the context of unverified claims about an internal OpenAI model, but no proof has been shown publicly. The one authority that matters for resolution, the Clay Mathematics Institute, has not acknowledged any submission or acceptance.
What Anthropic has actually verified
The rumor gained credibility because Anthropic’s recent, confirmed work is real. In August 2026, an unreleased Claude research model improved a lower bound linked to the Riemann hypothesis, raising the fraction of zeta zeros known to satisfy the hypothesis from 41.6% to 67.2%. According to TechCrunch and Anthropic’s own writeup, the run used roughly 60 subagents in Claude Code, two Anthropic mathematicians validated the result, and the proof was Lean-verified. That is a real improvement on a narrow technical bound, not a solution to the Riemann hypothesis itself.
Separately, Anthropic reported that Claude completed the first fully formalized, machine-checked Lean 4 proof of Fermat’s Last Theorem, writing what it describes as the largest Lean proof ever assembled over roughly 11 days. Importantly, that is formalization of an existing proof, following the Wiles argument, not the discovery of a new one. Neither result is a Navier-Stokes proof, and both are distinct from the unverified rumor.
| Date (2026) | Event | Status |
|---|---|---|
| August 2026 | Claude raises Riemann-zeta lower bound 41.6% to 67.2% | Verified by Anthropic mathematicians, Lean-checked |
| Early September | Claude formalizes Fermat’s Last Theorem in Lean 4 | Machine-checked formalization of an existing proof |
| September 5 | Andrew Curran predicts Claude solved Navier-Stokes before IPO | Unverified personal prediction, no source cited |
| September 6-8 | Terence Tao clarifies he knows of no breakthrough | Rumor unconfirmed; Clay problem still open |
| September 08 | Manifold “solved by 2027” market reads 79% “Yes” | ~50% of all-time volume traded in 24h, 99 bettors |
Why “considered solved” is a high bar
For the market to pay out “Yes” on June 3, 2027, several things have to happen in sequence. A proof would need to be public, survive sustained peer review in the partial differential equations community, and reach the level of consensus, ideally acknowledgment from Clay, that observers would recognize as “solved.” History counsels caution here. Claimed solutions to Navier-Stokes and other Millennium problems surface regularly and are usually found incorrect on scrutiny. A July 2026 preprint claiming a full proof of global regularity, for example, has not been endorsed by the wider community, and status trackers maintained through early September still describe the problem as open.
None of the three conditions is currently met. There is no public proof, no peer-reviewed publication that mathematicians broadly accept as solving the Clay formulation, and no acknowledgment from Clay of any prize submission. The “inside information” about a secret proof under review remains, on the public record, a rumor.
The broader tech and finance context
The reason a math problem is trading like a market-moving event is that it sits at the intersection of frontier AI and corporate timing. Curran tied his prediction to Anthropic’s expected IPO, which layers a financial narrative on top of a mathematical one. A verified Millennium Prize result attributed to an AI system would be a landmark for the field and a marketing moment for whichever lab produced it. That incentive structure is exactly why cautious traders demand hard evidence, preprints, conference talks, named expert endorsements, before assigning high probability, and why Tao’s public skepticism is meaningful negative signal.
This is not the first AI math question to draw a crowd on prediction platforms. Readers can compare the current frenzy with our coverage of the broader “AI solves a Millennium Prize problem” market and the more grounded case of the Fermat’s Last Theorem formalization odds, where the underlying achievement was concrete and verifiable.
What to watch next
Three signals would move this market on fundamentals rather than sentiment. First, any public statement from Anthropic, which has so far not confirmed a Navier-Stokes solution. Second, a preprint or formalized proof that named PDE specialists are willing to evaluate on the record. Third, any communication from the Clay Mathematics Institute, which is the reference authority the question ultimately depends on. Absent those, expect the price to stay reactive, sensitive to each new post and each clarification, and to drift back toward the base rate as the rumor ages without confirmation.
FAQ
Has Navier-Stokes actually been solved? No. As of September 08, 2026, the Clay Mathematics Institute still lists the existence and smoothness problem as open, and the $1 million prize is unclaimed.
Did Claude solve it? There is no public proof and no confirmation from Anthropic. The claim traces to one person’s prediction on X, framed as a prediction rather than a report.
What did Terence Tao say? He discussed a hypothetical scenario about AI-solved problems, then clarified he is aware of no confirmed breakthrough on Navier-Stokes and did not say Claude solved it.
What has Anthropic verified this year? A Lean-checked improvement to a Riemann-zeta bound (41.6% to 67.2%) and a formalized Lean 4 proof of Fermat’s Last Theorem. Neither is a Navier-Stokes result.
Is the 79% figure a real-money price? No. Manifold uses play money, so the number reflects attention and conviction more than capital at risk.
The Bottom Line
Sources
- Manifold — Will Navier-Stokes be considered solved on June 3th 2027?
- Clay Mathematics Institute — Millennium Prize Problems
- TechCrunch — An unreleased Anthropic model made progress on one of math’s biggest unsolved problems
- Anthropic — Learning more about Claude’s mathematical capabilities
- Anthropic — Formalizing Fermat’s Last Theorem
- explainx.ai — Did Claude Solve Navier-Stokes? The Rumor, Fact-Checked
- 36Kr — Did Claude Solve the Millennium Prize Problems? Terence Tao: I Never Said That
Prediction markets carry risk and are not investment or betting advice. Availability is restricted by jurisdiction: Polymarket is not available to US persons, and Kalshi is a CFTC-regulated US exchange. Manifold operates on play money. Age restrictions of 18+ or 21+ apply as applicable. If gambling is a problem for you or someone you know, call 1-800-GAMBLER for confidential help.
Related Coverage
- ARC-AGI-3 'Above Human' Bet Hits 92% on Manifold After GPT-6 Astra
- AI Index Clears 65: Traders Flood a Manifold Market After Claude Fable 5.1 Scores 66
- Harry-Meghan Divorce Bet Surges on Manifold as Couple Moves Back to the UK
- Fermat's Last Theorem Prediction Market Spikes to 99% After Claude's Lean Proof
- GTA VI 2026 Release Odds Hit 96% on Manifold After Netflix Extended Look