Why don’t teachers teach AI ethics? Understanding teachers’ beliefs and intentions in Chinese AI curriculum implementation through the theory of planned behaviour

While artificial intelligence (AI) education is expanding globally, the implementation of AI ethics education in K-12 AI curricula remains a critical challenge, particularly in top-down curriculum reform contexts like China. This study investigates the factors influencing secondary information technology (IT) teachers' intention to teach AI ethics, grounding in Theory of Planned Behaviour (TPB). Through semi-structured interviews with 14 in-service teachers in China's Greater Bay Area (GBA), we identified key behavioural beliefs (e.g., fostering critical thinking, navigating rapid technological change, and adapting western-centric ethics), normative beliefs (e.g., new curriculum policy, school-level expectations, and student interest), and control beliefs (e.g., deficient content knowledge, misaligned professional development, and classroom constraints). These factors interactively influence teachers' attitudes, social norms and perceived behavioural control, shaping their intentions to implement AI ethics education in their classroom practices. The findings reveal that while some teachers recognize the importance of teaching AI ethics, their intentions of implementing this domain are predominantly constrained by low perceived behavioural control and social norms that prioritizes technical aspects. This study advocates for collaborative professional learning where teachers develop competencies of AI ethics through group moral reasoning engaging with core ethical principles, which in turn enables them to reflect on pedagogical designs for creating active learning activities, thereby bridging the gap between policy commitment and classroom practice.