AI patients, real practice: exploring the use of AI-simulated patients to support primary healthcare training for combat medical technicians

INTRODUCTION: Combat Medical Technicians (CMTs) are central to military primary care but have limited opportunity for clinical exposure. Simulated patients offer a controlled method to maintain clinical currency. Advances in conversational artificial intelligence (AI) enable realistic and interactive simulated consultations. We present our evaluation of the feasibility, acceptability and educational impact of AI-simulated patients for CMT training. METHODS: Five military primary care simulated patients were developed and hosted on the SimFlow.ai platform and delivered during a development course. Participants completed pre- and post-simulation surveys assessing confidence across 12 clinical domains alongside perceptions of realism, usability and educational value. Quantitative analysis used Wilcoxon signed-rank tests and Spearman rank correlations. RESULTS: < 0.002). Evaluation of the simulations demonstrated positive perceptions of medical accuracy, patient narratives and overall educational value. Technical performance received mixed feedback, with response lags identified as the primary barrier. No associations were found between outcomes and CMT demographics which suggests equitable benefit across the cohort. CONCLUSION: AI-simulated patients are feasible to implement and are associated with meaningful improvements in consultation confidence among CMTs. Despite technical constraints, AI-simulated patients represent a scalable, standardised adjunct to support clinical currency across the CMT workforce. Further research should evaluate objective competence outcomes and explore broader uses of this technology.