OpenAI announces three new discoveries spanning mathematics, computer science, and physics, covering harmonic analysis, complexity theory, and quantum field theory. The manuscripts address the local smoothing conjecture related to the Kakeya conjecture, the Exponential Time Hypothesis in complexity theory, and quantum chromodynamics.
In response to academic community feedback, OpenAI is experimenting with a new release process. External experts were invited to shepherd each result, advising three distinct strategies: releasing raw AI output for the math result, forming a collaborator group for the CS result while keeping AI output public, and deferring to an external group for the physics result.
The external collaborations follow four principles: transparency (no NDAs), no academic credit sought by OpenAI, recognition for contributors, and independence of participants. For the CS collaboration, a natural language proof and Lean formalization have been published to GitHub. The physics group is withholding AI-produced artifacts to write up findings independently.
OpenAI acknowledges tradeoffs in selecting initial shepherds and welcomes feedback on broadening participation. These efforts aim to convert raw AI-generated research data into verifiable scientific understanding.
Source: OpenAI Blog · Summarized by HeadlinesBriefing