Answers are grounded in this paper and cite their sources.
Purpose of review Interest in autonomous surgery has intensified as artificial intelligence (AI) has progressed from preoperative applications to intraoperative perception, decision support, and robotic action. This review examines the current state of surgical autonomy and AI integration, with emphasis on urology. Recent findings Most cleared surgical robots remain at low levels of autonomy, but conditional autonomy has entered clinical practice. In urology, AI-guided Aquablation combines image-based treatment planning with robotic execution under surgeon authorization. In preclinical settings, autonomous systems have progressed from predefined tasks to learned control of soft-tissue manipulation, suturing, and multistep surgical actions. In parallel, surgical AI can recognize operative phases and gestures, assess technical performance, and provide real-time intraoperative decision support, with emerging multimodal and vision-language models further expanding these capabilities. Summary Surgical autonomy remains predominantly task-specific and human-supervised, with more advanced autonomous execution largely confined to preclinical settings. Near-term progress will likely involve increasing integration of conditional-autonomy functions and AI decision support into existing robotic platforms, with the surgeon retaining a central role in supervision and decision-making. Broader clinical adoption will require generalizable models, rigorous validation, and appropriate regulatory and accountability frameworks.
No takes yet. Share an insight, caveat, or question.
Zuluaga et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: