Artificial Intelligence and Surgical Education in the UK: A Systematic Review of Current Use, Evidence Gaps and Future Directions.

Artificial intelligence (AI) offers new opportunities to enhance surgical training through automated performance assessment, adaptive learning platforms, and AI-enabled virtual or augmented reality (VR/AR) simulation. Although global literature is expanding, the UK context differs in governance, procurement, and training structures. National initiatives, such as the Royal College of Surgeons of England (RCS) Future of Surgery (FOS) programme, have highlighted AI and extended reality as priorities for modernising surgical education. However, UK peer-reviewed evidence remains limited, with most work consisting of pilot studies and early feasibility assessments. This narrative review synthesises UK-specific applications of AI in surgical training, identifies current gaps, and proposes priorities for future research. Two independent reviewers conducted a focused search of peer-reviewed and grey literature, including RCS policy documents, to identify UK-based uses of AI in surgical training. Databases searched included PubMed, the Excerpta Medica database (Embase), Scopus, and Web of Science, supplemented by targeted screening of UK policy sources. Studies were included if they involved AI or AI-enabled technologies applied to surgical education, simulation, or assessment within the UK. Non-UK studies and articles focused solely on clinical (non-educational) AI applications were excluded. Data were synthesised on AI modality, educational outcomes, feasibility, and barriers. A Risk Of Bias In Non-randomized Studies - of Interventions (ROBINS-I)-aligned risk-of-bias assessment was performed.

resultsUK literature comprises national policy reports and a small number of empirical pilot studies exploring AI-enhanced VR/AR simulation, AI-driven performance analytics, and early AI components within robotic training curricula. RCS policy documents consistently identify AI as a key element of future training reform. Empirical studies report feasibility, trainee acceptability, and construct validity but provide limited evidence of improvements in operative performance or patient outcomes. Most work is single-centre and exploratory, and significant barriers, including cost, faculty training, data-governance requirements, and variability in access across deaneries, remain. DISCUSSION: The UK is at an early yet promising stage of adopting AI within surgical education. National policy momentum and the expansion of robotic programmes provide opportunities for coordinated integration. Collaboration between NHS education bodies, simulation centres, and technology developers could support standardised metrics and equitable access. However, robust multi-centre evaluation frameworks are required to determine educational effectiveness. Ethical considerations, including data privacy, algorithmic transparency, and the impact of automated feedback on trainee development, require careful attention.

conclusionsAI use in UK surgical training is emerging but currently driven largely by pilot studies and policy direction rather than high-quality outcome evidence. Major gaps include multi-centre validation, curriculum integration, standardised assessment frameworks, and equitable access to AI-enabled systems. Future UK research should prioritise structured validation studies and national coordination to define effective, scalable, and safe AI tools that can enhance surgical education across the NHS.