New AI Solves a Millennium Prize (thezvi.substack.com)

A groundbreaking development in AI and mathematics has emerged as a new model, enhanced by OpenAI, has successfully solved the Navier-Stokes problem—one of the seven Millennium Prize problems—after just eight days of training. This rapid progress has sparked both excitement and concern within the AI/ML community regarding the pace of AI advancements and its implications for future research. OpenAI emphasized the need for transparency about such rapid progress and the potential necessity for more deliberate management of AI development, given the involvement of significant computational resources, estimated to cost millions. The situation has been complicated by tensions between OpenAI and mathematicians Tristan Buckmaster and Levent Alpoge, who had been working on related mathematical proofs. They alleged that OpenAI had attempted to downplay their contributions and sought authorship changes that they found unacceptable. While OpenAI clarified that user data was unlikely to have influenced their results, concerns remain about the ethical implications of using AI in advanced mathematical research. This incident not only highlights the potential for AI to solve longstanding mathematical problems but also raises crucial questions about credit, collaboration, and the responsibilities of AI developers in academic discourse.