Current medical education is caught in a paradox: it aims to train compassionate patient advocates yet its structure often fosters compliance and burnout, ill-equipping physicians to challenge the systemic drivers of poor health. The rapid integration of artificial intelligence (AI) into clinical practice is automating the cognitive-diagnostic tasks that have long defined physician training, from pattern recognition to data synthesis. This shift presents a specific, time-limited opportunity: rather than simply adopting AI as an efficiency tool within the existing transactional model, medical education can use this moment to fundamentally reimagine the physician's role and the training that shapes it. This requires moving beyond traditional competency-based models to embrace a pedagogy of advocacy, grounded in critical consciousness and systems-level thinking. By transforming curricula to address structural determinants of health, reassessing how we evaluate competence, and creating protected space for advocacy, we can train a new generation of physicians prepared not only to treat disease but to partner in transforming the systems that perpetuate it, through a phased approach aligned with AI maturity in clinical practice.
Patel et al. (Fri,) studied this question.