OpenAI AI Agents Solve Navier-Stokes Singularity in 88 Hours

Summary

The Wall Street Journal segment reports that OpenAI's AI system solved the Navier-Stokes existence and smoothness problem—one of the seven Millennium Prize Problems—by demonstrating that smooth solutions can develop singularities in finite time. It highlights this as evidence of unexpectedly rapid AI advancement in mathematics, contrasting a 2000 Fields Medalist's view that such problems were inaccessible to computers with recent expert forecasts and the breakthrough. The piece weaves in concurrent events: an Anthropic researcher's public warning of >10% chance of human extinction from AI in the next decade, a colleague's resignation over safety concerns, and leaders from four major AI firms agreeing to slow development. It also notes OpenAI hosting a mathematicians' summit assuming superhuman math capabilities soon and shifting focus to human-solvable problems like education.

Sourcing relies on the OpenAI announcement, the Anthropic researchers' X posts, statements from Fields Medalists, and reporting on the summit and slowdown agreement. The narrative centers on a Fields Medalist expressing positivity alongside two dozen others fearing AI undermines math's purpose.

Editorial Assessment

The broadcast accurately captures a genuine, major mathematical milestone achieved with AI, with the Lean-formalized proof providing strong verification that the claimed singularity exists under the specified conditions. However, viewers miss critical context: OpenAI's effort followed rumors of progress by human mathematicians Tristan Buckmaster and Levent Alpöge (using AI tools from OpenAI and Anthropic), sparking accusations of possible influence or scooping; the precise formulation (forced Navier-Stokes with bounded energy) may not fully resolve every aspect of the Clay Millennium statement in the eyes of all experts. The segment's framing heavily emphasizes dread and loss of control, linking the math triumph directly to extinction risks and undermining math itself, which could skew perception toward viewing AI progress as primarily threatening rather than transformative. It omits that many mathematicians are excited by the potential for AI to accelerate discovery, even as 25 Fields Medalists signed a letter decrying the misalignment between AI companies' benchmarking goals and the field's emphasis on human insight. Overall, while facts check out, the selective emphasis and missing nuance on controversy and credit create a more alarmist tone than the evidence strictly supports.