This chapter discusses various aspects of AI agent safety and security, exploring potential vulnerabilities, mitigation strategies, and governance practices to ensure reliable and ethical AI systems. It discusses the risks posed by both accidental failures and deliberate attacks, highlighting the need for rigorous testing, redundancy mechanisms, and robust defenses against adversarial threats. Key considerations for inter-agent communication, embodied AI security, and ongoing governance are discussed. By addressing these challenges, the chapter aims to advance the safe deployment and management of AI agents.