So… did OpenAI actually just solve the Navier–Stokes problem?

I saw the news about OpenAI’s Navier–Stokes result couple days ago and at first I thought it meant AI had somehow solved an impossible problem. Turns out it’s a bit more complicated than that. The Navier–Stokes equations are already used to describe how fluids move. The famous unsolved part was basically this: if you start with a perfectly smooth 3D fluid, do the equations always stay well behaved forever, or can they eventually develop a singularity? OpenAI is claiming they found a proof that a singularity can form in finite time. According to OpenAI, they had groups of AI agents trying different approaches, sharing useful discoveries between them, and eventually produced both a mathematical proof and a formalized version in a coding language called lean. There’s obviously still the question of independent verification, and I’ve seen mathematicians being cautious about the problem is officially settled but if the proof holds up, I think the bigger story might actually be what this says about AI doing maths. If AI systems can start producing genuinely new proofs, what would doing a scientific research look like in 5-10 years period? Do mathematicians mostly become the people who choose the right problems, guide the models, and verify outcomes? Or is this being overhyped? Lmk what you think.