Efficiency vs. safety in AI-enabled medical education: an ethical analysis of AI as a bridge or a wedge

Abstract: Artificial intelligence is rapidly changing medical education, promising faster workflows and richer learning resources while quietly reshaping how future clinicians think and act. This paper examines the central tension between efficiency and safety in AI-enabled medical education, asking when AI functions as a bridge that strengthens training and when it becomes a wedge that undermines it. Drawing on a targeted review of 1,266 pieces of literature on clinical decision support, diagnostic algorithms, and generative AI, we identify three interlocking ethical tensions: clinical efficiency versus health equity, cognitive convenience versus clinical judgment, and data-driven personalization versus professional integrity. Using role conflict theory, we demonstrate how these tensions manifest in the daily work of learners and clinical educators, who must simultaneously prioritize patient safety, promote independent reasoning, and adapt to AI-mediated workflows. We argue that medical education should treat AI not only as a technical tool but as a curricular and ethical problem: learners must be trained to question, calibrate, and sometimes refuse AI outputs. Framed this way, the task is not to decide for or against AI, but to design conditions under which it reliably acts as a bridge rather than a wedge.