Background: Artificial intelligence (AI) is increasingly integrated into healthcare, creating a need for future physicians who should have the skills to use AI critically, responsibly, and professionally. Although AI education is expanding in undergraduate medical education, most evaluations focus on learner satisfaction or self-reported confidence, providing a limited understanding of how AI literacy develops during learning.

Aim: To explore the development of AI literacy among undergraduate medical students through longitudinal analysis of structured reflective logbooks completed during an AI course.

Methods: A qualitative longitudinal study was conducted using structured self-reflection logbooks completed after each session of a nine-day AI course at David Tvildiani Medical University (Georgia). Fifteen students enrolled, and logbooks from 12 students who completed the full longitudinal portfolio were analyzed. Data were examined using inductive qualitative content analysis, with codes iteratively developed and synthesized into higher-order themes representing the progression of AI literacy.

Results: Students' reflections demonstrated a developmental trajectory across four interconnected stages: Understanding AI, emphasizing foundational knowledge and recognition of AI capabilities and limitations; Understanding AI Systems, reflecting deeper understanding of data quality, machine learning, model evaluation, and generative AI limitations; Critical AI Use, characterized by prompt engineering, evidence-informed AI use, clinical applications, and critical appraisal of AI-generated information; and Professional AI Practice, focusing on ethics, governance, patient safety, professional accountability, and the continuing importance of human clinical judgment. Overall, reflections progressed from knowledge acquisition to critical, ethical, and professional integration of AI into healthcare.

Conclusions: Longitudinal reflective logbooks provided meaningful insight into AI literacy development during a structured AI course. The identified trajectory aligns with key domains of the UNESCO AI Competency Framework for Students, including AI techniques and applications, critical evaluation, ethical use of AI, and human-centered professional practice. Structured reflective logbooks may represent a valuable approach for evaluating AI literacy development in undergraduate medical education.