Comparison of artificial intelligence assisted training and traditional learning paths in clinical simulation skills training: meta-analysis of randomized controlled trials.

BACKGROUND: Artificial intelligence (AI) is increasingly integrated into clinical simulation training, yet its overall effectiveness compared to traditional learning remains unclear. This meta-analysis of randomized controlled trials (RCTs) aimed to evaluate the impact of AI-assisted training versus traditional instruction on clinical skills performance, procedure time, and learner satisfaction. METHODS: PubMed, Embase, and Web of Science were searched from inception to April 10, 2026. RCTs comparing AI-assisted training (real-time feedback, intelligent tutoring, AI-simulated patients, or image recognition support) with traditional instruction (expert-led, self-directed, or role-play) in simulation-based clinical skills training were included. The primary outcome was clinical skills score. Standardized mean differences (SMD) with 95% confidence intervals (CI) were calculated using fixed- or random-effects models based on heterogeneity. RESULTS: Twelve RCTs were included. Overall, significant difference was found between AI-assisted and traditional training for clinical skills score. Subgroup analysis revealed that AI significantly improved surgical and invasive procedural skills, but showed no advantage for clinical assessment and diagnostic skills. In addition, the vision-based deep-learning AI model may have more advantages in medical training than the traditional training method, while the large language model did not demonstrate statistically significant differences with the traditional training method. Procedure time and Operation completion rate did not differ between groups. Learner satisfaction and willingness to recommend the training and stress level of AI assisted training were higher than those of the control group. CONCLUSION: This study shows that AI, especially the vision-based deep-learning, can effectively guide the training of procedural skills. However, for complex cognitive skills like clinical reasoning, AI is best used to replace inefficient peer practice with standardized, high-intensity training. Expert teachers should then focus on advanced clinical judgment, emotional communication, and personalized correction. Overall, AI offers a major opportunity to improve quality and scale up medical simulation education. Its best path is deep integration with traditional teaching-combining their strengths rather than simply replacing teachers. SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/PROSPERO/view/CRD420261411402. The protocol was registered in advance in PROSPERO online platform as CRD420261411402.