The use of GenAI in medical education in China and Indonesia: A comparative literature review

Abstract

Background: The integration of artificial intelligence, particularly generative AI (GenAI), in learning is transforming medical education from undergraduate to subspecialty levels, promising enhanced diagnostic accuracy, personalized learning, and real-time feedback. However, concerns persist regarding reliability, accuracy, and ethical implications, including the potential for "hallucinations" and an erosion of students´ critical thinking skills. The evolving regulatory landscape also highlights challenges in ensuring safety, fairness, and accountability as GenAI systems gain autonomy in clinical decision-making. This narrative literature review examines the current landscape of GenAI applications in medical education within China and Indonesia, aiming to identify trends, gaps, and strategies for developing GenAI competencies among future healthcare professionals. Methods: A systematic search of PubMed, MedLine, Google Scholar, Scopus, and Chinese CNKI databases, along with targeted hand searches, identified 12 empirical studies published between January 2022 and December 2024. Studies were categorized based on Kirkpatrick’s evaluation levels. Results: We found that GenAI has been used across various medical topics and learning activities, including scientific presentations, small-group discussions, and simulations. Benefits included increased confidence in GenAI use, improved student engagement, and enhanced practical skill development. Conversely, challenges included sporadic adoption, lack of training, concerns about misinformation, technical limitations (e.g., accuracy in non-English contexts), and the high cost of implementation. Most evaluated outcomes were at Kirkpatrick's Level 1 (satisfaction) and Level 2 (knowledge/skills), with a notable absence of Level 3 (behavioral changes) or Level 4 (healthcare outcomes), suggesting the field is in early evaluative stages. Discussion and Conclusion: Our review recommends comprehensive training for faculty and students, curriculum integration, and robust evaluation systems to address accuracy and ethical concerns. Future research should focus on longer-term impacts on behavioral changes and patient outcomes, utilizing multi-methodological approaches and fostering interdisciplinary collaboration. This will ensure GenAI effectively complements, rather than replaces, human expertise in preparing competent and ethical healthcare professionals.