This paper addresses the problem of efficiently characterizing large-scale quantum systems, such as quantum analog simulators and megaquantum computers. This represents a significant challenge in quantum science, as the Hilbert space of a quantum system grows exponentially with system size. This paper highlights that recent advances in artificial intelligence (AI), which excels at high-dimensional pattern recognition and function approximation, have emerged as powerful tools for solving this challenge. Research on representing and characterizing scalable quantum systems using AI has been extensive, ranging from theoretical foundations to experimental implementations. These efforts can be categorized into three synergistic paradigms, including machine learning (especially deep learning) and language models, based on how AI is integrated. This paper discusses how each AI paradigm contributes to two core challenges in quantum system characterization: predicting quantum properties and generating surrogate models of quantum states. These challenges underpin diverse applications, ranging from quantum authentication and benchmarking to improving quantum algorithms and understanding the phases of strongly correlated matter. It also discusses key challenges, open issues, and future prospects for the interface between AI and quantum science.
