Lingzhi Bao, Jie Ma
The rapid evolution of artificial intelligence (AI) is driving a paradigm shift in modern healthcare, creating an urgent imperative to reform traditional medical education. This narrative review synthesizes the current landscape, core applications, challenges, and actionable pathways for responsible AI integration in medical education. AI enables personalized and adaptive learning, immersive clinical simulation, automated assessment and formative feedback, data-driven curriculum design, and efficient administrative support, while large language models (LLMs) further enhance knowledge synthesis, exam preparation, and clinical reasoning practice. However, widespread adoption remains constrained by infrastructure deficits and digital inequity, faculty AI literacy gaps, unresolved ethical concerns including algorithmic bias and data privacy risks, automation bias and potential skill degradation, and a critical shortage of rigorous longitudinal evidence on educational efficacy. This review argues that sustainable AI integration requires a shift from technological optimism to human-centered, evidence-based implementation that prioritizes human-AI collaboration, standardized governance, ethical stewardship, and equitable access. We conclude that AI, when thoughtfully deployed, can enhance competency cultivation while preserving the humanistic core of medical training, preparing future clinicians to practice safely and effectively in an AI-enabled healthcare ecosystem.