OpenAI's AI Math Breakthrough Sparks Fight Over Credit

OpenAI says its AI agents solved one of math's biggest mysteries: the Navier-Stokes problem. This puzzle asks whether the equations that describe how water and air flow can ever break down and predict something impossible, like infinite speed. It's one of seven Millennium Prize Problems, each worth one million dollars. Only one had ever been solved before. So this seems like a historic win for AI.

But the celebration has turned sour. Two mathematicians, Tristan Buckmaster and Levent Alpöge, say OpenAI used their nearly a year of work as a starting point without giving credit. They had been using public AI models from both OpenAI and rival company Anthropic. According to Buckmaster, OpenAI employees gave him two options: post their work first and let OpenAI follow, or collaborate on a paper that left Alpöge out because he works for Anthropic. OpenAI denies using their research unfairly.

The bigger picture matters more than this one dispute. If AI is now needed to solve the hardest problems in math, only big tech companies with massive computing power can participate. That breaks the old tradition of open academic collaboration. Buckmaster and Alpöge actually posted a simpler version of the proof they found on a social network, showing their method worked. Then OpenAI went further with the full problem. But the question of who deserves credit is now central.

This situation affects more than just mathematicians. It shows how AI is changing who gets to make discoveries and who gets recognized for them. When an AI tool builds on ideas from humans without clear rules, we need new agreements about fairness and credit. The math might be solved, but the ethical puzzle is just beginning.