Smarter, Faster, Safer: How AI Is Rewiring Sports, Performance Science, and Athlete Health

Background. Artificial intelligence (AI) is rapidly permeating sports medicine and performance science, propelled by high-volume data from wearables, video, and clinical systems. Beyond traditional analytics, modern machine learning (ML) and deep learning (DL) can model complex, nonlinear relationships to support diagnosis, injury prevention, load management, and tactical decisions. Yet adoption remains uneven due to concerns about data quality, bias, interpretability, governance, and the appropriate balance between automation and expert judgment. Aim. To synthesize emerging evidence on AI applications across sports medicine and performance optimization, clarifying opportunities, constraints, and ethical considerations required for safe, effective, and equitable integration. Material and methods. Narrative review of AI modalities, core data ecosystems and representative use cases spanning MSK imaging, injury-risk modeling, movement analysis, recovery optimization, and tactical decision support. The review emphasizes multimodal data fusion, model oversight, and real-world implementation factors. Current state of knowledge. Evidence indicates AI can enhance diagnostic speed and accuracy in MSK imaging, improve early detection of overuse and acute injury risk via spatiotemporal and physiological signals. enable individualized load management and adaptive training through time-series modeling; and support coaching and education via NLP and LLM-powered tools. Key limitations include dataset heterogeneity, biased sampling and gaps between prototype performance and field deployment. Conclusions. AI can make training and care more precise, proactive, and accessible, but benefits hinge on high-quality data pipelines, transparent models, clinician/coach oversight, and ethical safeguards. Interdisciplinary collaboration is essential to realize performance gains while protecting athlete welfare and equity.