Humanoid Robots Enter the Operating Room: A Complete Breakdown of the First Remote-Controlled Live Surgery
On July 8, 2026, the journal Nature published a study poised to reshape the surgical robotics industry: a team of engineers and surgeons at the University of California, San Diego (UCSD), successfully deployed two remotely operated humanoid robots in an operating room to conduct live surgery on large non-human primates. This marks the moment humanoid robots officially crossed the threshold from industrial inspection into the operating theater.
Experimental Design: Two Surgeries, Two Paradigms
The research team designed two contrasting experimental scenarios to validate different roles for humanoid robots in the surgical environment.
Surgery One — Human-Robot Collaborative Cholecystectomy. A humanoid robot nicknamed Surgie partnered with a human surgeon to form a surgical team. Surgie served as the primary operator, performing core steps — gallbladder dissection, ligation, and excision — under remote teleoperation, while the human surgeon acted as first assistant, retracting tissue and exposing the surgical field. This model simulates a real-world scenario involving one remote expert guiding a robot performing on-site.
Surgery Two — All-Robot Team Surgery. Two Surgie robots worked side by side — one as primary surgeon, the other as assistant — with no human surgeon directly handling instruments at any point. This means a surgical procedure could be completed without an on-site human surgeon, a capability with disruptive implications for medical rescue in remote areas and extreme environments.
Both procedures were performed on large non-human primates as a preclinical proof of concept. The ultimate goal is not to replace human surgeons but to build a "surgery room of the future" where humanoid robots and humans work together seamlessly.
The Technical Foundation: Why Humanoid Robots?
Current mainstream surgical robots — exemplified by Intuitive Surgical's da Vinci system — are highly specialized behemoths: weighing approximately 1,800 pounds (~816 kg), equipped with three to four dedicated robotic arms, proprietary tools, and custom software, they require retrofitted operating rooms and large teams for deployment.
In contrast, the Surgie used in this study — adapted from Unitree's G1 humanoid platform — stands 5 feet (~1.52 m) tall and weighs only 60 pounds (~27 kg). The researchers developed custom adapters enabling Surgie to grip conventional laparoscopic instruments. This form-factor difference yields advantages across three dimensions:
First, deployment flexibility. Surgie can walk directly into a standard operating room without structural modifications or dedicated installation teams. Dr. Shanglei Liu, assistant professor of surgery at UCSD School of Medicine and co-senior author, stated: "It is a fraction of the cost. It is a fraction of the real estate taken in the operating room. So it is easy to deploy, whether that be in a rural setting or up in space."
Second, multi-task generality. Dedicated surgical systems can only perform pre-programmed, procedure-specific tasks. A humanoid robot, by contrast, can walk, retrieve objects, and clean the operating room, and in the future could prepare instruments pre-operatively and handle sterilization post-operatively. Professor Michael Yip, professor of electrical and computer engineering at UCSD Jacobs School of Engineering and senior author, noted: "One of our goals is to develop autonomous surgical assistants. Many communities face surgical team staffing shortages, and humanoid robots could serve as integrated team members to bridge that gap."
Third, intuitive control. For medical personnel not trained on da Vinci systems, controlling a humanoid robot feels more akin to natural human motion. Dr. Nikita Thareja, general surgery resident at UCSD School of Medicine and study co-author, admitted: "We were surprised at how well Surgie fit into our workspace and our workflow."
Precision and Challenges: From Proof of Concept to Clinical Deployment
On surgical precision, Dr. Liu confirmed: "A surgery performed by a remotely controlled humanoid robot is as precise as one performed by a remotely controlled surgical robotic system." This refutes the intuitive assumption that lightweight platforms necessarily compromise precision.
Yet significant challenges remain. During surgery, the robot required multiple recalibrations, resulting in substantially longer procedure times compared to existing dedicated surgical systems. Dr. Liu remains optimistic: "The first laparoscopic robotic surgery took six hours; today it takes 30 minutes. Humanoid surgical robots will follow a similar learning curve."
Latency in teleoperation represents another critical bottleneck. The delay between a surgeon's hand-controller movement and the robot's execution increases with distance. The team is actively optimizing the control system for long-distance telesurgery scenarios, targeting sub-perceptual latency levels.
Industry Implications: Unitree G1's Dual Identity
Notably, the robots used in this experiment are Unitree's G1 humanoid robot, originally released in 2025. A platform designed as a "general-purpose humanoid" transformed into a surgical primary operator once equipped with UCSD's custom end-effector adapters. This reveals a deeper trend: the value of humanoid robots lies not in purpose-built specialization but in the ability to cover multiple verticals through swappable end-tooling — from warehouse material handling to surgical instrument manipulation, all sharing the same embodied intelligence core.
Future Roadmap: From OR to Battlefield to Space
Professor Yip outlined a clear three-phase roadmap for Surgie: near-term (surgical assistant) → mid-term (remote teleoperated surgery) → long-term (autonomous surgical assistant). This trajectory extends beyond conventional hospital settings — the paper specifically highlights surgical needs in search-and-rescue field medicine and space stations, environments where dispatching a human surgeon is neither practical nor economical.
Dr. Ryan Broderick, interim director of the UCSD Center for the Future of Surgery, concluded: "This achievement exemplifies the power of bringing together engineer and surgeon innovators to solve meaningful clinical problems in our world-class training and research laboratory. Our center bridges engineering innovation with clinical expertise, empowering transformative ideas to be rigorously developed, tested, and refined."
From an 1,800-pound specialized giant to a 60-pound general-purpose humanoid surgical assistant, the "slimming revolution" of surgical robotics is officially underway. When a Unitree G1 picked up a scalpel and stepped onto the operating table, what we witnessed was not just a paper publication — it was the birth of an entirely new track in medical robotics.