Artificial Intelligence in Orthopaedic Training: A Narrative Literature Review of Applications, Evidence, and Limitations.

Orthopaedic surgical training has traditionally relied on an apprenticeship model, but reduced working hours and heightened patient safety expectations have narrowed the operative exposure available to trainees, prompting growing interest in adjuncts to conventional training. Artificial intelligence (AI) is increasingly being applied within surgical education as one method of supporting learning, offering new approaches to simulation, technical skills assessment, and personalised learning. Given the rapid pace of development in this field, this narrative review synthesises the current evidence for AI applications in Trauma & Orthopaedics training, examining three key domains: AI-enhanced simulation, objective skills assessment, and predictive analytics for personalised learning pathways. The evidence indicates consistent, if preliminary, benefit from AI-enhanced virtual reality (VR) simulation, including improved arthroscopy skills and reduced operative duration, alongside promising applications of haptic feedback for real-time coaching and personalised, algorithmically generated performance feedback. Objective, motion-tracking-based skills assessment significantly improved with established tools such as Objective Structured Assessment of Technical Skills (OSATS) and Global Evaluative Assessment of Robotic Skills (GEARS), though this evidence largely originates from general and robotic surgery rather than orthopaedics specifically. Predictive analytics offer early potential to forecast trainee learning curves and identify those at risk of underperformance but raise ethical concerns regarding premature labelling and its effect on trainee confidence and progression. Across all domains, evidence remains limited by small, single-institution studies, a lack of randomised controlled trials, and a near-total absence of research linking AI-assessed simulation performance to long-term patient outcomes. Cost and unequal access to high-fidelity simulation technology have the potential to further constrain widespread adoption. AI therefore holds considerable promise for orthopaedic training but should currently be regarded as a complement to, rather than a replacement for, traditional apprenticeship-based education, pending stronger longitudinal and comparative evidence.