OpenAI keeps bulldozing mathematicians

Your ongoing mathematical research may become irrelevant overnight, as AI models are poised to announce solutions to many more problems soon.

OpenAI has formed an independent advisory group of elite mathematicians called AGMAI to help review and communicate its emerging mathematical results, aiming to repair its relationship with the community after past announcement missteps. The group, including prominent figures like Martin Hairer, was announced via a blog post by UCLA professor Terence Tao. Despite OpenAI emphasizing AGMAI's independence, the announcement caused confusion, with some mathematicians initially believing it was an OpenAI-appointed body. AGMAI members, while stressing their autonomy and lack of financial ties to OpenAI, acknowledged that the company's initial approach was the impetus for its formation. The group is already facing the daunting task of helping coordinate the release of numerous new results from OpenAI's unreleased models, which have reportedly solved over 100 long-standing mathematical problems. This rush to release, coupled with past issues like poorly written manuscripts and lack of proper credit, has generated concern and dread within the mathematical community.

The formation of AGMAI, while intended to foster trust, has been marred by OpenAI's chaotic announcement style, creating confusion about the group's true independence. This has led mathematicians to question OpenAI's commitment to collaborative research practices and proper scholarly conduct, highlighting a persistent disconnect between AI development timelines and academic norms. The situation underscores the challenge frontier AI labs face in integrating their rapid progress with the established, deliberate pace of scientific discovery and peer review.

OpenAI's current strategy suggests a significant, potentially disruptive, influx of AI-generated mathematical solutions is imminent, creating anxiety among researchers about the devaluation of their work and the potential for shoddy scholarship. The company's approach, characterized by a focus on quantity and speed over detailed understanding and credit, signals a burgeoning burden on the mathematical community to validate and contextualize these findings. This dynamic raises questions about the future of mathematical progress and whether AI will primarily serve as a tool for discovery or a force that preempts human effort.