Modern scientific challenges—from predicting complex fluid flows to modeling plasma behavior in fusion reactors—demand computationally efficient, trustworthy surrogates that can rival traditional numerical solvers while harnessing the power of artificial intelligence. Scientific machine learning (SciML), identified as a core technology for AI, offers immense potential for surrogate modeling in both data‑rich and data‑scarce situations; of particular interest is the field of operator learning. However, current operator learning frameworks lack unified theoretical foundations, robustness guarantees, and scalable training methods, which limit their adoption in high-stakes applications. The Unified Neural Operator (UNO) considered in this project will fill this gap by embedding all operator learning techniques into a unifying framework, marrying the mathematical rigor of traditional methods with the expressivity of modern AI. By delivering certifiable, interpretable AI‑driven surrogates, UNO advances Presidential priorities in artificial intelligence and nuclear energy—supporting both next‑generation AI capabilities and efficient modeling of magnetohydrodynamic systems critical for fusion energy—while fulfilling NSF’s mission “to promote the progress of science; to advance the national health, prosperity, and welfare; and to secure the national defense” Within SciML, operator learning has shown tremendous potential as a powerful tool for creating surrogate models, leading to a