OpenAI’s GPT-5.2 has done something remarkable: it derived a new formula in theoretical physics that human scientists later verified. The discovery concerns gluons, the particles responsible for the strong nuclear force that holds atomic nuclei together. For decades, physicists believed a particular type of gluon interaction — involving one gluon with negative spin and the rest with positive spin — was simply impossible, resulting in a zero amplitude. The new preprint shows that under specific conditions called the “half-collinear regime,” this interaction actually does occur. The formula was first proposed by GPT-5.2 Pro, with human researchers working out the complex calculations to confirm it. This isn’t just a mathematical curiosity; it suggests AI can now make meaningful contributions to fundamental physics by identifying patterns humans might miss. The work also has implications for graviton interactions, potentially bridging quantum mechanics and gravity. Why it matters: This marks a shift from AI as a calculator to AI as a genuine collaborator in theoretical discovery, raising questions about how much of future physics breakthroughs might come from machine intuition.