Artificial Intelligence in Physical Education: A Systematic Review of Personalized Learning, Assessment, and Performance Analytics

The purpose of the study. This systematic review examines how artificial intelligence (AI) is applied to personalized learning, assessment, and performance analytics in physical education (PE) across K–12 and higher-education settings, with the aim of synthesizing empirical evidence, identifying patterns of implementation, and proposing evidence-based directions for future research and practice. Materials and methods. A systematic review was conducted following the PRISMA 2020 guidelines. Seven electronic databases (Web of Science, Scopus, EBSCOhost, PubMed, ACM Digital Library, Taylor extracted data; and appraised quality using the Mixed-Methods Appraisal Tool (MMAT). Results. A total of 87 studies (from an initial pool of 2,945 records) met all inclusion criteria and were synthesized narratively. AI-based systems most commonly supported: (a) personalized learning through adaptive exercise plans and intelligent tutoring systems; (b) assessment via motion analysis and automated feedback mechanisms; and (c) performance analytics through wearable-driven dashboards and learning-analytics platforms. Overall, AI-enhanced PE was associated with improved student engagement, more accurate and objective assessment, and tailored motor-skill development. However, persistent concerns included data privacy vulnerabilities, algorithmic bias, and insufficient frameworks for teacher–AI collaboration. Conclusions. AI holds substantial potential to transform PE into a more personalized, data-informed, and student-centered discipline, particularly in large-class and inclusive settings. Future research should prioritize longitudinal designs, standardized outcome measures, and robust ethical frameworks to ensure equitable and sustainable integration of AI in PE contexts.