This paper presents a hybrid quantum-classical AI model for ethical decision-making training in medical education. Leveraging Qiskit and a neural network to parameterize a 2-qubit circuit, the system simulates ethical dilemmas using 860 synthetic patient scenarios focused on autonomy, consent, and resource allocation. It evaluates decisions based on probabilistic outcomes for trust and health. Compared to a classical rule-based baseline, the quantum model achieves higher performance—62% win rate for trust, 78% for health—demonstrating its effectiveness across diverse dilemma types. Visual analysis highlights how quantum computing better reflects real-world ambiguity. While based on simulated data, this proof-of-concept reveals strong potential for integrating quantum AI into medical training. It offers an interactive, data-driven framework that prepares students for complex clinical ethics through experiential learning. Connecting quantum cognition with educational goals, the model makes it possible to proceed with future validations using real-world data and deployment on physical quantum hardware.