Researchers from Stanford and Princeton universities have unveiled MedOS, a novel artificial intelligence system designed to serve as a real-time co-pilot for clinicians in operating rooms and hospital settings. Announced on February 11, 2026, by the Stanford-Princeton AI Coscientist Team, MedOS combines smart glasses, robotic arms, and multi-agent AI to perceive clinical environments in 3D, reason through medical scenarios, and assist with procedures. The system aims to reduce medical errors, accelerate precision care, and support overburdened clinical teams, addressing physician burnout that affects over 60% of doctors in the United States.
MedOS builds on the team's previous work with LabOS, bridging digital diagnostics with physical action. It has been tested in surgical simulations, hospital workflows, and live precision diagnostics. The system introduces a "World Model for Medicine" that creates a continuous feedback loop of perception, intervention, and simulation. Using smart glasses and robotic arms, MedOS can interpret real-time video, identify anatomical structures, and assist with robotic tool alignment, functioning as an active collaborator in high-stakes procedures rather than a passive assistant.
Key capabilities include a multi-agent AI architecture that achieved 97% accuracy on MedQA (USMLE) and 94% on GPQA, outperforming frontier models like Gemini-3 Pro and GPT-5.2 Thinking. The system also features MedSuperVision, the largest open-source medical video dataset with over 85,000 minutes of surgical footage from 1,882 clinical experts. In studies, MedOS helped nurses and medical students reach physician-level performance, reducing human error in fatigue-prone environments: registered nurses improved from 49% to 77% accuracy, and medical students from 72% to 91%.
Case studies demonstrated MedOS's ability to uncover immune side effects of the GLP-1 agonist Semaglutide (Wegovy) from the FDA database and identify prognostic implications of driver gene co-mutations on cancer patients' survival. The platform is modular and designed to adapt across clinical settings and specialties. MedOS is launching with support from NVIDIA, AI4Science, and Nebius, and early pilots have been deployed. Clinical collaborators can request early access through the project page at https://medos-ai.github.io/.
Dr. Le Cong, leader of the Stanford-Princeton AI Coscientist Team and Associate Professor at Stanford University, stated, "The goal is not to replace doctors. It is to amplify their intelligence, extend their abilities, and reduce the risks posed by fatigue, oversight, or complexity. MedOS is not just an assistant. It is the beginning of a new era of AI as a true clinical partner." Dr. Mengdi Wang, co-leader of the collaboration, added, "MedOS reflects a convergence of multi-agent reasoning, human-centered robotics, and XR interfaces. Our goal is a collaborative loop that helps clinicians manage complexity in real time."
MedOS will be showcased at a Stanford event in early March 2026, followed by a public unveiling at the NVIDIA GTC conference in March 2026. Session information is available online at https://www.nvidia.com/gtc/session-catalog/sessions/gtc26-s81748/. More information can be found on the official site at https://ai4medos.com/.
