The application and practice of artificial intelligence in physical education and training

Abstract

This study systematically investigates the application and effectiveness of Artificial Intelligence (AI) in collegiate Physical Education and Training (PET) through a mixed-methods design. The experimental group employed AI tools including computer vision-based motion analysis, wearable fitness trackers, and machine learning-driven personalized platforms, while the control group received conventional instructor-led training. Quantitative results revealed that the experimental group achieved significantly greater improvements in technical action standardization (25.3% mean score increase), 1000m run time (28.5 seconds reduction vs. 12.3 seconds), and standing long jump distance (15.2cm vs. 6.7cm increase). Student satisfaction was markedly higher in the AI-assisted group (4.52±0.38 vs. 3.21±0.45). Qualitative analysis of interviews with 10 instructors and 20 students identified key themes: enhanced assessment objectivity, personalized training adaptability, and practical barriers such as equipment cost and technical complexity. The study provides empirical evidence and practical insights to support the integration of AI in PET.