JupiteX Get the app
Science & Technology21 Nov 2026 · about 7 min

Unforced Navier-Stokes Singularity and the Generalized Riemann Hypothesis

The brief

The article reports three results involving AI-assisted mathematics: finite-time blowup proofs for the three-dimensional unforced incompressible Euler and Navier–Stokes equations, plus a Lean-only proof of the Generalized Riemann Hypothesis. The author orders them by how fully the results have been digested, not by importance. Levent Alpöge, Tristan Buckmaster, and Matei Coiculescu have written a human-readable Euler proof and formalized it in Lean. Alpöge and Buckmaster’s Navier–Stokes proof is also in Lean and its broad strategy is clear: build an approximate discretely self-similar solution numerically, then use rigorous perturbation theory to obtain an exact one. The author says this paper needs a more professional write-up. Alpöge’s enormous Lean-only GRH proof is at an earlier stage of human understanding. A Lean Zulip thread is tracking verification and attempts to extract its strategy, which initial analysis suggests draws on Langlands and arithmetic geometry. The author expects digestion to take time; moderators are now focusing that discussion on Lean-related aspects. The article does not report a completed independent verification.

01

What results does the article report, and how far along is the checking of each proof?

The article reports three results involving AI-assisted mathematics: finite-time blowup proofs for the three-dimensional unforced incompressible Euler and Navier–Stokes equations, plus a Lean-only proof of the Generalized Riemann Hypothesis. The author orders them by how fully the results have been digested, not by importance.

Levent Alpöge, Tristan Buckmaster, and Matei Coiculescu have written a human-readable Euler proof and formalized it in Lean. Alpöge and Buckmaster’s Navier–Stokes proof is also in Lean and its broad strategy is clear: build an approximate discretely self-similar solution numerically, then use rigorous perturbation theory to obtain an exact one. The author says this paper needs a more professional write-up.

Alpöge’s enormous Lean-only GRH proof is at an earlier stage of human understanding. A Lean Zulip thread is tracking verification and attempts to extract its strategy, which initial analysis suggests draws on Langlands and arithmetic geometry. The author expects digestion to take time; moderators are now focusing that discussion on Lean-related aspects. The article does not report a completed independent verification.

02

In a fluid equation, what does it mean for a solution to develop a singularity or “blow up” in finite time?

A singularity, or blowup, in a fluid equation is a mathematical failure of a solution’s regular behavior at a finite time. The solution may be defined and smooth up to a time T, but some quantity needed to keep it regular can grow without bound as time approaches T.

For example, the velocity itself might become unbounded, or its spatial derivatives might become unbounded. Which quantity matters depends on the precise theorem and the equation’s notion of a regular solution. A blowup result does not necessarily say that every feature of the fluid becomes infinite; it says that the solution can no longer meet the stated regularity conditions.

The article uses “finite time blowup” for claimed results about three-dimensional unforced incompressible Euler and Navier–Stokes equations. It does not specify which quantity becomes singular or give the precise hypotheses on initial data. Those details would be needed to state the exact mathematical result, so the explanation here gives the standard meaning rather than filling in theorem-specific claims.

03

What is the Generalized Riemann Hypothesis, and what does it say about the zeros of certain mathematical functions?

The Generalized Riemann Hypothesis, or GRH, is a family of conjectures about zeros of certain complex functions called L-functions. In a standard formulation, it says that every nontrivial zero of the relevant functions has real part one-half. The “generalized” version extends the idea beyond the original Riemann zeta function.

These zeros encode information about number patterns, so their locations have consequences for results throughout number theory. The exact family of functions depends on which formulation of GRH is being discussed. The source article does not define the hypothesis or identify its functions; this description is standard mathematical background, not a detail supplied there.

The article reports that Levent Alpöge used an internal AI model to generate an enormous Lean-only proof. It says initial analysis suggests the proof uses modern work on the Langlands program and arithmetic geometry, but emphasizes that people still need to verify it and extract its strategy. The report alone does not establish that the proof is correct.

04

How broad is a blowup result: does it show that some initial fluid conditions can lead to a singularity, or that every flow does?

A blowup result is usually an existence statement: it produces or establishes at least one suitable initial fluid state whose evolution becomes singular in finite time. It does not normally say that every possible initial state, or every flow, must blow up. Many flows may remain regular; an existence theorem needs only a qualifying example.

For the reported Navier–Stokes result, the article describes a strategy that numerically builds an approximate discretely self-similar solution and then uses rigorous perturbation theory to turn it into an exact solution. That points to constructing a particular solution, rather than asserting that all solutions behave this way. The article also reports an Euler blowup proof, but gives no comparable details about its construction.

The source does not state the exact assumptions, function spaces, or initial conditions in either theorem. So the existence-versus-every-flow distinction is the standard interpretation of a blowup claim, while the full scope of these particular claims must be checked in their proofs. The article presents them as results whose proofs still require digestion and checking.

05

What does formalizing a proof in Lean allow mathematicians to check, and why might a Lean proof still be difficult for people to understand?

Lean is a proof assistant: a mathematician expresses definitions and logical steps in a formal language, and Lean checks that the proof follows from the stated assumptions and its accepted rules. This can make a proof’s formal inferences machine-checkable. It does not automatically establish that the theorem was encoded as intended or that every assumption is appropriate.

The article reports Lean formalizations for both fluid-equation results and an enormous Lean-only GRH proof. A formalization may contain many technical steps and unfamiliar notation. Even when Lean accepts it, readers may struggle to see the main idea, why the definitions match the intended mathematics, or how the argument can be presented in ordinary mathematical prose.

That distinction appears directly in the report. The Euler result has a human-readable proof as well as a Lean formalization. The Navier–Stokes paper is technically human-readable, but needs a more polished proof. For GRH, the author says people are still working to verify the proof and extract its strategy. Lean checking and human understanding are related, but not the same task.

06

If these proofs withstand expert checking, what would they establish for mathematics?

If the proofs withstand expert checking, they would establish the specific mathematical claims they prove, rather than merely suggest them through computation or AI output. In particular, the fluid results would establish finite-time blowup for the stated unforced, incompressible equations in three dimensions. The GRH result would establish the Generalized Riemann Hypothesis in the formulation covered by its formal statement.

The Navier–Stokes strategy described in the article illustrates the difference between a numerical clue and a proof: numerical work builds an approximate discretely self-similar solution, then rigorous perturbation theory is used to obtain an exact one. A correct proof would turn that route into a mathematical result. The article also reports a human-readable Euler proof alongside its Lean formalization.

These would be consequential claims, especially because the GRH concerns a central conjecture in number theory. But the article presents verification and explanation as ongoing work, most explicitly for the GRH proof; the Navier–Stokes paper also needs a more professional write-up. This answer is conditional: the report itself does not establish that every proof has passed expert checking.

07

What do the Euler and Navier–Stokes equations model, and how do incompressibility, viscosity, and forcing shape the fluid motion they describe?

The Euler and Navier–Stokes equations are mathematical models for how fluids move. Incompressibility means the modeled fluid does not locally expand or compress; mathematically, its velocity field has zero divergence. The article specifically concerns incompressible equations in three dimensions.

The key difference is viscosity: Navier–Stokes includes a term representing internal friction that spreads and smooths motion, while Euler is the idealized inviscid model without that term. Forcing means an external input, such as an imposed force, drives the fluid. The article calls both reported blowup results unforced, so it is not describing externally driven motion in these claims.

These features determine which mathematical equations are being studied, and a blowup result concerns whether their solutions can lose regularity. The article reports proofs for the unforced incompressible Euler and Navier–Stokes equations in three dimensions. It does not explain the physical setting or state the equations’ precise terms, so these definitions are standard background rather than further details from the report.

This brief was written by AI from the original reporting and checked by other models. Names, figures and quotes come from the source; read it for full context.

Read more in the JupiteX app

Pulse is free. New stories every 4 hours, each one broken into the questions that explain it.

Or read more news on the web