Abstract: Drawing upon Maxwell’s equations as its theoretical underpinning, computational electromagnetics has achieved notable progress in high-power electromagnetic domains, encompassing areas like electromagnetic forecasting and fault diagnosis. In recent years, with the development of artificial intelligence (AI) technology, its powerful data fitting and nonlinear learning capabilities have garnered widespread attention in the field of high-power electromagnetic computation. This has opened up new possibilities for addressing several research bottlenecks in high-power electromagnetics, such as data limitations and large computation. This paper provides an overview of various research findings on the application of AI in high-power electromagnetic computation, and the development of this field is prospected.