Orthodontists' Perceptions, Attitudes, and Adoption of Artificial Intelligence: A Scoping Review

Background: Artificial intelligence (AI) is increasingly utilized in orthodontics for diagnosis, treatment planning, and clinical management. Nevertheless, considerable variation persists in orthodontists' knowledge, attitudes, utilization patterns, recommendations, and concerns regarding AI. The aim of this scoping review was to synthesize survey-based evidence on orthodontists' perceptions and experiences with artificial intelligence. Methods: A systematic literature search was performed in PubMed, Embase, Scopus, and Web of Science to identify survey-based studies evaluating AI awareness, applications, attitudes, recommendations, and concerns among orthodontists, residents, postgraduate students, and academicians. Results: Seven studies involving a total of 1772 participants were included. Overall, postgraduate students and practicing clinicians demonstrated relatively limited knowledge of AI, whereas academicians exhibited a higher level of awareness. Although routine clinical implementation of AI remains limited, it was most frequently applied-or perceived as beneficial-in cephalometric and cone beam computed tomography (CBCT) analysis, orthognathic surgery planning, and treatment outcome prediction. The majority of participants supported the promotion of AI and the integration of AI education into orthodontic curricula. However, concerns persisted regarding insufficient technical expertise, high costs, ethical and legal challenges, reduced patient engagement, and the potential for diagnostic or procedural errors. Consistent variations related to age and professional role were observed across studies among academicians, postgraduate students, orthodontists, and residents. Conclusions: This scoping review seems to suggest a growing interest in and generally positive attitudes toward AI among orthodontists and trainees. However, the evidence base is limited to a small number of studies, heterogeneous and predominantly based on cross-sectional survey data. For these reasons, findings should be interpreted cautiously. Variability in knowledge and use persists, and integration into practice remains inconsistent. Further research is needed to support effective and evidence-based implementation.