A heated academic and tech debate has emerged after OpenAI published what it claims is a solution to the Navier‑Stokes problem, a notoriously difficult math challenge with a $1 million prize attached.
NYU professor Tristan Buckmaster had announced progress using AI models, including OpenAI’s, only to see the company unveil its own breakthrough less than 24 hours later.
Anthropic Link Adds Tension to AI Research Drama
Buckmaster’s post detailed tensions with OpenAI, including claims the company suggested coordinating announcements and omitting his collaborator, Levent Alpöge, who works at rival Anthropic.
OpenAI researcher Sebastien Bubeck denied this, saying he never asked for Alpöge’s removal. OpenAI acknowledged it began working on the problem after learning of Buckmaster’s research but insisted its models did not access user data.
The episode raises broader concerns: if you use AI tools to develop ideas, could their owners front‑run you?
Apple’s lawsuit against OpenAI has already highlighted risks of trade secrets being absorbed into AI training. OpenAI CFO Sarah Friar has even suggested the company could profit when users’ AI‑enabled work generates revenue.
For academics and entrepreneurs alike, the saga underscores the need to understand how AI platforms handle intellectual property and user contributions.
