Artificial Intelligence and Robotics in General Surgery: Opportunities and Challenges

Artificial intelligence and robotics are reshaping general surgery across the full perioperative continuum. This narrative review traces the history of surgical robotics from early teleoperated systems developed by NASA and the US Department of Defense to the current da Vinci Xi and SP platforms and examines how AI is now being applied at each phase of surgical care. In the preoperative setting, machine learning models outperform traditional risk scores in predicting postoperative complications, while deep learning applied to computed tomography (CT) and magnetic resonance imaging (MRI) improves tumor detection, lymph node staging, and surgical planning. Intraoperatively, AI-driven phase recognition systems achieve 85-95% accuracy in identifying procedural steps, computer vision tools assess the Critical View of Safety during laparoscopic cholecystectomy, and semi-autonomous robotic systems are beginning to reduce surgeon tremor and automate discrete operative tasks. In the postoperative period, AI-integrated wearable biosensors and electronic health record models enable earlier detection of complications such as sepsis and deep vein thrombosis, while personalized ERAS protocols are refined through continuous data streams. Despite these advances, significant challenges remain, including dataset bias, limited external validation, black-box model opacity, and unresolved questions around data privacy, liability, and regulatory oversight. AI currently functions as a decision-support tool rather than an autonomous actor. Broader clinical adoption will require larger multi-institutional datasets, improved model interpretability, and clear regulatory frameworks.