Integrating Large Language Models and Personalized Learning inMedicalEducation: Potential, Challenges, and the Path Ahead – ASystematicReview

Abstract

The use of large language models (LLMs) in medical education, particularly for personalized learning, has gained significant interest in the past five years. We conducted a systematic literature review (2019–present) following PRISMA guidelines to assess how LLMs (e.g., GPT-based models) are applied in adaptive medical training. Searches of PubMed, Scopus, IEEE Xplore, and Web of Science identified relevant peer-reviewed articles, prioritizing systematic reviews, meta-analyses, and qualitative studies. LLMs have the potential to transform medical education through adaptive quizzes, on-demand tutoring, and customized study plans. However, challenges such as misinformation, algorithmic biases, and privacy concerns must be addressed. We outline guidelines to leverage LLM benefits while ensuring faculty oversight to maintain accuracy. If responsibly integrated, LLMs can significantly enhance personalized learning in medical education. Collaboration among educators, AI developers, and policymakers is essential to establish safe, effective practices. This review synthesizes current evidence, identifies innovation opportunities, and highlights necessary safeguards for AI-driven medical education.