GPT-5.2 derives a new result in theoretical physics

We’ve published a new preprint showing that a type of particle interaction many physicists expected would not occur can in fact arise under specific conditions. The work focuses on gluons, the particles that carry the strong nuclear force. The preprint⁠(opens in a new window) is available on arXiv and is being submitted for publication. In the meantime, we welcome feedback from the community.

The preprint, titled “Single-minus gluon tree amplitudes are nonzero,” is authored by Alfredo Guevara (Institute for Advanced Study), Alex Lupsasca (Vanderbilt University and OpenAI), David Skinner (University of Cambridge), and Andrew Strominger (Harvard University), and Kevin Weil (OpenAI) on behalf of OpenAI.

The preprint studies a central concept in particle physics called a scattering amplitude. A scattering amplitude is the quantity physicists use to compute the probability that particles interact in a particular way. For gluons, the particles that carry the strong nuclear force, many amplitudes take unexpectedly simple forms “at tree level” (meaning calculations that keep only the simplest diagrams with no quantum loops). These simplifications have repeatedly revealed deeper structure in quantum field theory, the framework that provides a description of physics that unifies special relativity with quantum mechanics.

One case, however, has generally been treated as absent (having zero amplitude). When one gluon has negative helicity (meaning one of the two possible spin orientations a massless particle can have) and the remaining n−1 n-1 gluons have positive helicity, standard textbook arguments suggest that the corresponding tree-level amplitude must be zero. As a result, this configuration has largely been set aside.

With the help of GPT‑5.2, these amplitudes have already been extended from gluons to gravitons, and other generalizations are also on their way. These AI-assisted results, and many others, will be reported on elsewhere.