The success of artificial intelligence (AI) applications in public health depends on its domain-specific data in public health (e.g., its relevance and structure). It begins with AI pipeline perspectives and introduces Electronic Medical Records (EMR) and the “Meaningful Use” (MU) framework for available health data. The chapter then systematically explores the distinct data needs for two key application areas: (1) systemic-factor public health AI, including disease surveillance, epidemiological forecasting, emergency response, population health analytics, Social Determinants of Health (SDOH), resource management, and policy optimization; and (2) personal-responsibility public health AI, which focuses on data concepts and metadata necessary for personal health records and individualized interventions. By categorizing the domain data, this chapter serves as a vital guide for AI practitioners and public health experts to identify, acquire, and structure the domain-specific data necessary to train robust and impactful AI models. Those concepts for domain data are the foundation for public health students to understand and explore various AI applications in public health.