Generative AI in physical education and health: a narrative review and conceptual framework for interdisciplinary thematic learning.

Generative artificial intelligence (GenAI) has created new possibilities for teaching and learning, yet its pedagogical role in physical education and health (PEH) remains underdeveloped. Current AI-related research in PEH has largely focused on performance-oriented and technical applications, such as motion analysis, skill evaluation, and training monitoring, with less attention to curriculum design, interdisciplinary learning, and teacher-mediated pedagogical use. This narrative review synthesizes literature on GenAI in PEH and related interdisciplinary or thematic learning contexts to examine how GenAI may support interdisciplinary thematic learning in PEH. The review highlights an emerging shift from AI as a technical or analytical tool toward GenAI as a pedagogical resource for lesson planning, assessment design, feedback, inquiry support, and knowledge integration. Based on this synthesis, the study proposes a pedagogical framework that positions GenAI as a teacher-mediated resource for connecting bodily practice, health knowledge, reflective inquiry, and social participation. The framework includes four core dimensions: curriculum and goal alignment, content organization and task design, inquiry and learning process support, and assessment and adaptive adjustment. It also emphasizes human-AI collaboration, teacher judgment, embodied observation, and ethical use as necessary conditions for meaningful implementation. The review suggests that GenAI may support PEH when it is integrated carefully into pedagogically designed, context-sensitive, and health-promoting learning processes. Future empirical research, particularly design-based studies and classroom interventions, is needed to test and refine the framework in authentic PEH settings.