Artificial intelligence for teaching, training, and assessment in dental education: a domain-based scoping review

Background: Artificial intelligence (AI) is rapidly transforming dental education by enhancing preclinical skill development, clinical diagnostic training, assessment processes, and content generation. Despite increasing interest, the scope and methodological characteristics of AI integration across dental curricula remain unclear. This review aimed to map current applications, benefits, and challenges associated with AI in dental education. Methods: Following the Arksey and O'Malley framework and PRISMA-ScR guidelines, a systematic search was conducted across major databases in December 2025. Seventeen empirical studies met the inclusion criteria. Data were charted using a structured extraction tool and synthesized descriptively. Studies were categorized into four thematic domains: preclinical training, clinical and diagnostic training, assessment and feedback systems, and AI-generated educational content. Methodological characteristics and commonly reported limitations (e.g., sample size, outcome type, comparator presence, and validation approach) were mapped descriptively to contextualize the evidence. Results: AI demonstrated promising applications across domains, including improvements in procedural accuracy, diagnostic consistency, assessment workflows, and learning material generation. However, the evidence base was heterogeneous and frequently limited by small sample sizes, short evaluation periods, reliance on self-reported outcomes, and limited external validation. Key gaps included limited real-time procedural assessment and insufficient educator involvement in AI design. Conclusion: AI offers substantial opportunities to enhance dental education but requires standardized definitions, stronger methodological rigor, ethical governance, and improved faculty readiness. Clinician-led, collaborative AI development will be critical to ensuring safe, pedagogically aligned integration.