Perceptions, Trust, and Barriers Related to the Adoption of Artificial Intelligence for Assessing Dental Implant Success and Failure Among Dental Professionals: A Cross-Sectional Survey.

Introduction Artificial intelligence (AI) is increasingly being explored in implant dentistry to improve diagnosis, treatment planning, and the prediction of implant success or failure. However, successful clinical implementation depends largely on dental professionals' awareness, trust, and acceptance. This study aimed to evaluate the awareness, knowledge, perceptions, trust, acceptance, and barriers to the adoption of AI for assessing dental implant success and failure among dental professionals. Materials and methods A questionnaire-based cross-sectional survey was conducted among 450 dental professionals, comprising 150 practitioners, 150 academicians, and 150 trainees. The validated questionnaire contained 15 items covering four domains: demographic characteristics, awareness and knowledge, perceptions and attitudes, and trust and acceptance of AI. Descriptive statistics were calculated, and intergroup comparisons were performed using the chi-square and Kruskal-Wallis tests. Mann-Whitney U tests, Jonckheere-Terpstra trend analysis, and multivariable ordinal logistic regression were performed where appropriate. Statistical significance was set at p < 0.05. Results Significant differences in AI awareness and acceptance were observed among the professional groups. Awareness of AI applications in implant assessment was highest among academicians, followed by trainees and practitioners (p < 0.001). High or complete trust in AI-generated implant predictions was reported by 85 (56.7%) academicians, 70 (46.7%) trainees, and 40 (26.7%) practitioners. Similarly, willingness to incorporate AI into routine clinical practice was significantly greater among trainees and academicians than among practitioners (p < 0.001). Lack of training and education was the most frequently reported barrier, whereas practitioners primarily identified implementation costs, and academicians highlighted ethical and medicolegal concerns. Multivariable ordinal logistic regression demonstrated that trainee status (odds ratio (OR) = 7.03), academic status (OR = 3.49), greater AI familiarity (OR = 2.27), and higher trust (OR = 2.59) were significant positive predictors of AI adoption, whereas increasing age (OR = 0.64) and years of experience (OR = 0.92) were negatively associated with willingness to adopt AI. Conclusions Dental professionals demonstrated significant differences in their awareness, trust, and acceptance of AI for implant success and failure assessment. Greater familiarity and trust were associated with greater willingness to adopt AI, whereas older age and greater clinical experience were associated with lower acceptance. Structured educational initiatives, clinical validation, and appropriate regulatory frameworks may facilitate the integration of AI into implant dentistry.