Robotic surgery (RS) is considered a stage in the evolution of minimally invasive procedures. Despite the increasing adoption of RS, challenges remain regarding outcome reproducibility, pronounced operator dependence, learning curves, and technological and organizational constraints, which underscores the need to further develop the “digital layer” of surgery and to apply artificial intelligence (AI) for intraoperative data analysis. A systematic literature review was conducted using Russianand English-language keywords related to RS and AI. The review summarizes key directions of AI integration into robot-assisted procedures, most commonly implemented as modules for surgical video and instrument kinematic analysis, including computer vision, phase and task recognition, extraction of objective performance metrics, automated skill assessment and training support, predictive modeling of selected postoperative course parameters, and elements of partial automation of subtasks while maintaining the primacy of clinical oversight. The interpretation of published results is limited by heterogeneity in datasets and evaluation protocols, as well as differences in annotation strategies, performance metrics, and validation approaches, which complicates crossstudy comparisons and model transferability across institutions. Clinical implementation of AI is further associated with requirements for data privacy and security, cyber-resilience, model transparency, and regulatory accountability.