You should care about the AI math breakthrough drama even if you're not a nerd
The math community is heated over questions about credit and AI training data after a mathematician's allegations against OpenAI went viral.
The AI research community is abuzz with controversy after two mathematicians claimed to have made breakthrough progress on a notoriously difficult mathematical problem using AI tools. Tristan Buckmaster, a mathematician at NYU, and Levent Alpöge, a researcher at Anthropic, published papers detailing their findings, but their claims have sparked debate over scientific credit and the potential misuse of AI training data.
The problem they tackled is known as the Navier-Stokes Millennium Prize Problem, a question first proposed in 1912 that seeks to understand the behavior of fluids like water and air. Solving this problem could earn the researchers a $1 million prize. While Buckmaster and Alpöge did not solve the problem outright, their work has been hailed as "remarkable" by renowned mathematician Terence Tao.
What set this story apart is the alleged involvement of AI companies OpenAI and Anthropic. Buckmaster alleges that OpenAI learned of their findings and told him that an internal AI model had made progress on solving the problem. OpenAI has since claimed to have independently solved the Navier-Stokes Millennium Prize Problem, but Buckmaster has not seen the proof and questions whether OpenAI used their data or chats to train its models.
The situation raises complex questions about scientific credit and the role of AI in research. Buckmaster argues that he and Alpöge used AI models from OpenAI and Anthropic to assist in their research, but he is unsure if OpenAI used their data or chats. OpenAI maintains that no specific user data was accessed to solve the problem, but it cannot rule out the possibility that de-identified data derived from user usage helped improve their models.
The dispute has sparked a debate about the ethics of using AI in scientific research and the potential for AI companies to train their models on groundbreaking work by human researchers. OpenAI's terms of service state that the company is allowed to train on user-submitted data unless users opt out, but Buckmaster claims that OpenAI was not clear about whether their AI models had been trained on their work.
As the controversy unfolds, the online math and AI communities are left to grapple with the implications of this high-stakes drama.
Written by urgent.news from Business Insider's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
