understanding and preventing ai-facilitated cheating in dental education

Background: generative ai has created new academic-integrity risks in professional training. in dentistry, these risks carry direct patient-safety implications.method: we conducted a structured narrative review (sanra-guided) of pubmed, scopus, and google scholar (september 2022–july 2025). original studies, reviews, and expert opinions on ai-facilitated academic misconduct in dental education were eligible; clinically focused ai papers were excluded. dual screening and thematic synthesis were applied.results: sixteen studies met the inclusion criteria; ~19% were dental-specific, with the remainder informing transferable practices from health/allied higher education. three themes emerged: (1) prevalence & patterns—≈58.7% of students report awareness of ai use by peers; applications include essay generation, clinical note fabrication, image manipulation, and real-time exam assistance; (2) detection challenges—traditional plagiarism tools detect ≤23% of ai-generated text; faculty report high uncertainty; (3) emerging solutions—authentic/oral assessments (≈73% reduction in cheating reports), policy frameworks, faculty development, explainable/authorship-verification tools, and integrity culture initiatives.conclusion: ai-facilitated cheating requires discipline-specific responses that combine policy (explicit acceptable-use definitions), pedagogy (authentic/oral/practical assessments), and platforms (fit-for-purpose detection/verification). priorities include clarifying policy, upskilling faculty, and validating detection approaches in clinical assessment contexts.