The rapid growth of artificial intelligence (AI) in public health needs robust evaluation frameworks. This chapter provides an overview of evaluation methodologies specifically targeted at AI applications in the public health domain. It begins with foundational understandings of general evaluation concepts (e.g., quality, safety, and usability) and approaches (e.g., objectivist and subjectivist). Moving beyond theory, the chapter presents a practical, phased framework detailing what to evaluate across the entire AI lifecycle: from pre-project feasibility and during design/development to post-deployment monitoring of system functions, user interactions, and ultimate public health outcomes. The chapter provides case studies that examine “the accuracy of AI diagnostic tools,” “the accountability of autonomous systems,” and “the user experience of AI-driven mental health support.” This chapter equips public health students, researchers, practitioners, and policymakers with the critical knowledge to design rigorous evaluations in AI for public health.