In the past decade, Artificial Intelligence (AI), as an important branch of computer science, has made breakthroughs in the research fields of computer vision, natural language processing, machine translation and so on. As a disruptive technology in the early 21st century, AI was rapidly applied to various research fields, including earthquake science. However, although the AI technology represented by machine learning and deep learning obviously exceeds the traditional algorithm in terms of performance, the model structure is usually much more complex. The nature of the black box and the lack of transparency hinder the decision-level application of AI technology in most research fields. In this context, explainable AI technology came into being, which aims to create a new or improved set of AI technology to help human users create a new generation of AI system that can be understood and trusted. This article firstly introduces the definition and methods of explainable AI, then exemplifies the application research of AI technology in earthquake science, and then discusses future development trend of explainable AI technology. Finally, the application prospect of explainable AI technology in earthquake science is proposed.
