Here is how to understand OpenAI’s major mathematical breakthrough — SkimNews
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- OpenAI used more than 10,000 AI agents working together in different configurations to show that the Navier-Stokes equations can 'blow up,' producing a solution in about 88 hours and resolving one aspect of the $1 million Millennium Prize Problem.
- The blow-up describes a spinning whirlpool that grows longer, thinner, and more intense over time until its swirling becomes infinite in finite time, using only a finite amount of force on the fluid.
- Luis Martínez-Zoroa (CUNEF University, Spain) and Diego Córdoba (Spanish National Research Council) laid the groundwork in 2021 with a thin-layer method for creating blow-ups, but their approach relied on an unrealistic force.
- Tristan Buckmaster (NYU) and Levent Alpöge (Anthropic) published a breakthrough hours before OpenAI that removed the unrealistic-force requirement for Euler equations, paving a clear path to Navier-Stokes.
- John Evans (University of Colorado Boulder) said the 166-page proof distills down to 'a couple of simpler physical mechanisms' an undergraduate could learn, and that the AI's trick was ultimately distinct from prior approaches, not just a more powerful repetition.
- Steven Brunton (University of Washington) called AI 'absolutely pivotal' but emphasized 'the solution approach was developed by brilliant humans,' and said the equations remain useful for engineers despite the blow-up finding.
- Zaher Hani (University of Michigan) warned that as AI proofs arrive thick and fast, 'it's not clear if we will get the same long-term benefits to the scientific process, or our mathematical understanding.'
Why it matters: OpenAI's proof demonstrates that large-scale AI agent collaboration can crack open problems that have resisted human mathematicians for decades — but the article makes clear the breakthrough depended on human-developed mathematical frameworks (Martínez-Zoroa/Córdoba in 2021, Buckmaster/Alpöge this week), and mathematicians are now racing to understand whether AI-generated proofs deliver genuine insight or just faster answers.
Ask SkimNews
