The last decade has seen an explosion of applications for generative AI methods in fundamental physics. At its core, this AI sub-field learns to efficiently estimate and sample from complicated probability distributions. And recently, generative AI has proven to be successful across many domains of human knowledge.

Due to both its impressive performance and universality, generative AI has been broadly adopted to help meet the growing need for complex simulations of big-data in high energy and condensed matter physics. In the pursuit of next-generation discoveries, both of these physics domains will increasingly rely on generative AI to meet simulation demands. Scientific simulations, however, require assurances of uncertainty quantification and interpretability; aspects which are comparatively lacking in current generative AI methods. Establishing trust in these methods therefore necessitates guarantees of robustness and a certain level of interpretability.

Experts spanning high energy and condensed matter physics, computer science, and industry are parallelly engaged in developing generative AI methods for physics simulations, each contributing their unique perspectives and techniques. The goal of this program is to bring together experts from these communities to work towards a shared objective: the development of effective, robust, and interpretable generative AI methods to simulate big data in physics applications. This highly interactive program will spark collaborations to develop novel methods and further our theoretical understanding of generative AI in physics applications.