Health Information Management (HIM) students are required to possess a high level of competence in understanding and applying disease and procedure coding based on ICD-10 and ICD-9-CM. However, conventional learning methods that are less engaging and unresponsive have resulted in limited understanding of the correct coding logic among students. This study aims to develop a chatbot-based learning model as an innovation to enhance students’ comprehension of diagnosis coding. The research employed a Research and Development (R&D) approach using the ADDIE model to produce an interactive and effective chatbot-based learning system for improving diagnostic coding skills. The study population consisted of 37 fifth-semester HIM students who had completed the Diagnosis Coding course. Data were collected through observation, interviews, questionnaires, and pretest–posttest assessments. The research instruments included media and material validation questionnaires based on learning media development standards, as well as a conceptual understanding test that had been tested for validity and reliability. Data were analyzed descriptively, both quantitatively and qualitatively, by comparing students’ performance before and after using the chatbot to assess its effectiveness. The media validation results showed a feasibility score of 91% (very feasible), and the material validation reached 89% (feasible). The average pretest score of 63.2 increased to 84.7 in the posttest after learning with the chatbot. Furthermore, 92% of students responded positively to the media, finding it engaging, easy to use, and helpful in understanding diagnostic coding. The chatbot-based learning model proved effective in enhancing students’ comprehension of diagnostic coding and has the potential to serve as an adaptive digital learning innovation in the era of health information system transformation .