This book introduces the concept and scope of AI safety and analyzes the security issues involved in AI as well as their governance strategies and solutions from different perspectives such as theoretical technology, technical standards, the industrial ecology, and laws. In addition to the overview, this book describes the methods of using trusted computing technology to solve AI safety problems, as well as the safety of unmanned systems, strong AI based on brain - like computing and its safety, the application safety of AI in intelligent manufacturing and smart cities, network security, the construction of an industrial ecosystem for AI safety and controllability, AI safety standards and risk assessment and early warning. Overall, this book not only outlines the general attributes of AI safety issues, but also analyzes the special security problems in AI systems, as well as their risk assessment, prediction, governance approaches and solutions from different perspectives and in different typical application fields.