Artificial intelligence (AI) is increasingly applied across all disciplines of medicine, including dentistry. Oral health research is experiencing a rapidly increasing use of machine learning (ML), the branch of AI that identifies inherent patterns in data similarly to how humans learn. In contemporary clinical dentistry, ML supports computer-aided diagnostics, risk stratification, individual risk prediction, and decision support to ultimately improve clinical oral health care efficiency, outcomes, and reduce disparities. Further, ML is progressively used in dental and oral health research, from basic and translational science to clinical investigations. With an ML perspective, this review provides a comprehensive overview of how dental medicine leverages AI for diagnostic, prognostic, and generative tasks. The spectrum of available data modalities in dentistry and their compatibility with various methods of applied AI are presented. Finally, current challenges and limitations as well as future possibilities and considerations for AI application in dental medicine are summarized. The amount of electronic health data generated globally is sharply increasing. In dental medicine, this data can be divided into three major overarching modalities: image data, structured numerical data, and unstructured textual data. Image data include radiography, photography, ultrasonography, near-infrared light transillumination, histology, and three-dimensional point clouds. Structured numerical data include demographic data, electronic dental records, as well as clinical parameters. Unstructured textual data include free-form clinical notes, patient and vendor correspondence, journal articles, and other written material. Diagnostic modeling typically involves recognizing structures and patterns in the input data or classifying these into largely predefined categories. In dental medicine, these input data are predominantly visual, with structured numerical data playing a secondary role. Imaging methods require visual assessment by the observer; with increasing complexity, the diagnostic difficulty increases. In addition, human diagnostic accuracy is inherently a function of the observer's experience and attention, as well as the prevalence of the diagnosis. To mitigate these limitations, computer-aided diagnostics use ML and computer vision to aid the human diagnostic process. Prognostic modeling uses various algorithms to forecast outcomes based on a set of inputs. Discriminative models, or models that learn P(Y|X), are particularly effective for classification tasks with discrete outputs. These models use training data to learn the boundaries between classes and then utilize the learned boundaries to predict a class, or a forecast, from a given input. Regression models are also widely used for prognostic modeling, particularly for continuous outputs. Generative models learn the probabilistic distribution of the input space P(X) and use these probabilities to generate new data. Often in response to prompts, generative models create synthetic new data with similar characteristics to their training data. Numerous generative models each employ a distinct method for content creation. The most commonly used approaches include generative adversarial networks (GANs), transformer-based models, and diffusion models. AI has transformative potential in dental research and clinical care. For clinicians, application of AI in dentalArtificial intelligence (AI) is increasingly applied across all disciplines of medicine, including dentistry. Oral health research is experiencing a rapidly increasing use of machine learning (ML), the branch of AI that identifies inherent patterns in data similarly to how humans learn. In contemporary clinical dentistry, ML supports computer-aided diagnostics, risk stratification, individual risk prediction, and decision support to ultimately improve clinical oral health care efficiency, outcomes, and reduce disparities. Further, ML is progressively used in dental and oral health research, from basic and translational science to clinical investigations. With an ML perspective, this review provides a comprehensive overview of how dental medicine leverages AI for diagnostic, prognostic, and generative tasks. The spectrum of available data modalities in dentistry and their compatibility with various methods of applied AI are presented. Finally, current challenges and limitations as well as future possibilities and considerations for AI application in dental medicine are summarized. The amount of electronic health data generated globally is sharply increasing. In dental medicine, this data can be divided into three major overarching modalities: image data, structured numerical data, and unstructured textual data. Image data include radiography, photography, ultrasonography, near-infrared light transillumination, histology, and three-dimensional point clouds. Structured numerical data include demographic data, electronic dental records, as well as clinical parameters. Unstructured textual data include free-form clinical notes, patient and vendor correspondence, journal articles, and other written material. Diagnostic modeling typically involves recognizing structures and patterns in the input data or classifying these into largely predefined categories. In dental medicine, these input data are predominantly visual, with structured numerical data playing a secondary role. Imaging methods require visual assessment by the observer; with increasing complexity, the diagnostic difficulty increases. In addition, human diagnostic accuracy is inherently a function of the observer's experience and attention, as well as the prevalence of the diagnosis. To mitigate these limitations, computer-aided diagnostics use ML and computer vision to aid the human diagnostic process. Prognostic modeling uses various algorithms to forecast outcomes based on a set of inputs. Discriminative models, or models that learn P(Y|X), are particularly effective for classification tasks with discrete outputs. These models use training data to learn the boundaries between classes and then utilize the learned boundaries to predict a class, or a forecast, from a given input. Regression models are also widely used for prognostic modeling, particularly for continuous outputs. Generative models learn the probabilistic distribution of the input space P(X) and use these probabilities to generate new data. Often in response to prompts, generative models create synthetic new data with similar characteristics to their training data. Numerous generative models each employ a distinct method for content creation. The most commonly used approaches include generative adversarial networks (GANs), transformer-based models, and diffusion models. AI has transformative potential in dental research and clinical care. For clinicians, application of AI in dental