Use of artificial intelligence as an adjunctive tool in the diagnosis of oral potentially malignant disorders: A literature review

Objective: To analyze the available scientific evidence regarding the use of Artificial Intelligence (AI) as an adjunctive tool for clinical evaluation in the diagnosis and screening of oral potentially malignant disorders (OPMDs), thoroughly assessing its diagnostic reliability, technical limitations, and relevant ethical aspects. Methodology: This study is a literature review in which structured searches for articles published between 2011 and 2026 were conducted in the PubMed/MEDLINE, SciELO and Google Scholar electronic databases, using controlled descriptors and Boolean operators related to AI and oral pathology. Results: The scientific literature consistently demonstrates that Deep Learning models, trained primarily on conventional intraoral clinical photographs, achieve high sensitivity and accuracy rates in the detection of OPMDs. However, significant challenges remain, such as high methodological heterogeneity among studies, reliance on well-standardized training data, and the need for more diverse population representation. Conclusion: Artificial Intelligence emerges as a highly promising adjunctive technology for screening and clinical decision support. Nevertheless, its application must remain strictly complementary, in no way replacing the clinical judgment of the dental professional or the histopathological gold standard obtained through tissue biopsy.