In-class AI usage and university students' attitudes: The mediating role of perceived learning engagement benefits.

AI is increasingly embedded within health systems worldwide, reshaping how care is delivered, managed, and organized [-]. The World Health Organization (WHO) defines AI as the “ability of algorithms encoded in technology to learn from data so that they can perform automated tasks without every step in the process having to be programmed by a human” []. These systems include a broad range of approaches such as machine learning methods and large language models. AI-based applications are now used across clinical and nonclinical settings to support risk prediction, patient triage, workflow management, and population health surveillance [,], influencing both direct care and broader health service organization. Global investment in AI for health continues to rise rapidly, projected to reach US $125 billion by 2028 [], reflecting strong institutional commitment to its integration into health systems. However, the impact of these technologies is not determined by investment or technical capability alone [,]. Their adoption and effectiveness in practice depends fundamentally on how they are understood, interpreted, and used by health care workers in real-world clinical and organizational contexts [-]. In this context, AI literacy has emerged as a key component of digital competence in health care [,]. AI literacy refers to the knowledge, skills, and attitudes required to understand how AI systems function [,], critically evaluate their outputs [], and apply them appropriately within professional practice [,]. It also encompasses ethical awareness and the ability to engage with the organizational and sociotechnical contexts in which AI is deployed [,]. Rather than being a fixed competency, AI literacy is increasingly understood as an adaptive capacity that must evolve alongside rapidly changing technologies and clinical applications []. Despite its growing importance, empirical evidence on AI literacy among practicing health care workers remains limited. Much of the existing literature focuses on conceptual frameworks, position statements, or policy guidance, with limited attention to measurable competencies in practice [,,]. At the organizational level, AI readiness indices primarily assess infrastructure, governance, and institutional capacity, but provide limited insight into individual-level competence []. Yet, it is this self-assessed competence that ultimately determines whether AI tools are used safely, effectively, and appropriately in clinical decision-making [].