AI
OpenAI's Race for Mathematical Glory Is Fracturing the Research Community
Mathematicians accuse the AI company of prioritizing competition over collaboration as it chases Millennium Prize problems.
The Verge
OpenAI's recent claim of solving one of mathematics' most famous unsolved problems—the Navier-Stokes equation—should have been cause for celebration. Instead, it has exposed deep rifts between how tech companies and academic mathematicians approach their work.
The Breakthrough and the Backlash#
The company deployed roughly 10,000 agents and invested tens of millions in computational resources to solve the Navier-Stokes problem in 88 hours. This represented a genuine technical achievement, but many researchers view it less as a scientific milestone and more as a demonstration of brute-force competitive advantage. The underlying tension: mathematicians pursue understanding and discovery; OpenAI appears driven primarily by the desire to win.
At the center of the controversy are two researchers—Tristan Buckmaster at NYU and Levent Alpöge at Anthropic—who were independently working on the same problem. According to Buckmaster's account, OpenAI learned of their progress and accelerated its own effort. What followed, he claims, was an offer that crossed an ethical line: OpenAI researcher Sébastien Bubeck allegedly offered Buckmaster virtually unlimited computational resources and sole authorship of the company's paper—but only if he distanced himself from Alpöge, whose affiliation with rival Anthropic created a conflict.
Buckmaster rejected the offer outright, calling it a "bribe." He also raised concerns about whether his own use of OpenAI's Codex tool (which generates code) might have inadvertently informed the models behind the company's solution. OpenAI has denied that user data influenced its work, though it initially could not rule out that derivative information from his prompts contributed to improvements in the system.
A Different Philosophy#
The philosophical gap between academic research and corporate ambition runs through this dispute. Mathematics, in its traditional sense, values the lineage of ideas—how techniques and methods developed to solve one problem often prove more valuable than the solution itself. These innovations ripple through the field, shaping new research directions for decades. Mathematicians care deeply about credit not for ego, but because acknowledging the conceptual lineage enables the field to advance.
Tech companies, by contrast, measure success through the problems themselves. Bubeck's surprise when Buckmaster said he didn't care about the Millennium Prize itself—a $1 million award—underscores this gulf. In Silicon Valley, the prize and the prestige are the point. In academia, they are sidelines to the pursuit of truth.
