Artificial intelligence (AI) has enormous potential in the field of structural failure, particularly for predicting, detecting and preventing failures in various infrastructures and other critical structures. This study explores the application of AI methods in the field of reservoir dams to simulate slip failure scenarios. Starting with second-level reliability methods based on first-order Taylor approximations (FOREM), using artificial intelligence (AI) techniques such as machine learning (ML) and deep learning (DL), tools are developed to simulate the effects of different load factors and stresses by taking the variation of physical characteristics at the dam-foundation interface and contact surfaces. To determine the safety status of a dam, the conventional artificial neural network (ANN) method has the advantage of being able to learn complex models from large amounts of data. The aim of the study is to establish a link between the failure probabilities Pf, calculated using the FOREM method, and the various elements associated with the functional safety of the dam. The elements influencing sliding include the rate of operation of the drainage system, the impact of the level of silting on the dam, the operational efficiency of the sluice gates and evacuation structures in the event of flooding, and the influence of the presence or deterioration of the injection veil. The aim is to determine the limits of the values associated with a probability of dam failure that approaches the probable class. This information provides engineers with solid lessons and justifications for better scheduling of maintenance operations and management of the hazards associated with dam failure, helping to reduce operating costs and boost public safety.