Generative Artificial Intelligence in University Sports Training: A Mixed-Methods Study

Abstract

This study investigates the application of Generative Artificial Intelligence (GenAI) in university sports training programs. A mixed-methods approach was employed, combining a quasi-experimental design with qualitative interviews. A total of 120 student-athletes from track and field and basketball programs were divided into experimental (GenAI-supported training) and control (traditional training) groups. Quantitative data included pre-and post-intervention performance metrics (e.g., speed, accuracy, endurance), while qualitative data were gathered through semi-structured interviews with 10 coaches and 15 athletes. Results indicated that the experimental group showed statistically significant improvements in performance outcomes compared to the control group (p < 0.05). Thematic analysis revealed that GenAI was perceived as highly effective for personalized training plan generation, real-time technique feedback, and motivational support. However, challenges included data privacy concerns, over-reliance on technology, and the need for specialized trainer upskilling. The findings suggest that GenAI has substantial potential to enhance sports training in higher education settings, but its integration must be pedagogically sound and ethically guided. This study contributes to the growing body of literature on AI in physical education and provides practical implications for sports educators and policymakers.