First reported Sep 17 — we wrote this up later than the original.
Humanoid Robots Assist Surgeons in Seoul Operating-Room Trial
Two humanoid robots handed tools and held a laparoscope alongside human doctors in a simulated gallbladder surgery in Seoul.
On September 16 at the Samsung Life Ilwon Station Building in Gangnam-gu, Seoul, a surgical team made up of human doctors and two humanoid robots reenacted a cholecystectomy — gallbladder removal surgery — in a demonstration organized by Samsung Medical Center, as reported by News1.
Instead of a human patient, a goat's liver was placed on the operating table while a laparoscopic camera fed a real-time view of the gallbladder and surrounding blood vessels to the surgical team. One humanoid robot took on the role typically played by a scrub nurse, retrieving and receiving surgical instruments on voice command — responding to instructions like "Give me the scissors" and "Take the tool." The second humanoid acted as an assistant surgeon, holding the laparoscope steady and retracting tissue as the lead surgeon called out adjustments such as "Closer" and "A little deeper." The lead surgeon then dissected the gallbladder, clipped and ligated blood vessels, and removed it using the same steps as an actual operation.
The robots are the product of the ORchestra (Operating Room Collaborative Humanoid) project, developed under South Korea's Ministry of Health and Welfare-backed Korean ARPA-H program. Samsung Medical Center leads the initiative, with participation from Rainbow Robotics, Aidin Robotics, Seoul National University, Sungkyunkwan University, the National Cancer Center, and Jeonbuk National University Hospital.
Turning the OR into a data factory
Rather than simply programming robotic arms, the research team is building what it calls a "surgical data factory." Ceiling-mounted cameras capture the movements of medical staff, wearable sensors record fine hand and finger motion, and laparoscopic footage is broken into "action triplets" — data structures linking the surgical scene, the instrument used, and the surgical step. Four institutions, including Samsung Medical Center, are currently contributing standardized data, which is anonymized after participant consent before being used to train AI models.
Otolaryngology professor Jung Yong-gi of Samsung Medical Center said the humanoid system converts the movement of a surgeon's hands and arms into a robot coordinate system so it can execute matching actions in the OR. The team has already published research on an algorithm that predicts the next surgical step and required instrument from laparoscopic video, with the longer-term goal of having robots prepare tools before the surgeon even asks. Jung noted the technology isn't yet able to capture the subtle technique differences between individual doctors, and said a full contextual "foundation model" of the operating room is still needed.
Comparisons to global efforts
Asked about Elon Musk's remark last year that AI would replace surgeons within three years, Jung called it "not an unfounded statement," framing it as a symbolic, directional claim rather than a literal timeline. He pointed to last year's SRT-H (Surgical Robot Transformer-Hierarchy) study published in Science Robotics, in which AI performed gallbladder removal on excised pig organs, and a July study from UC San Diego published in Nature showing a teleoperated humanoid performing surgery on a live pig. Jung argued that the Seoul demonstration — a teleoperated robot working alongside two humanoids as human staff directed them — represents a world-first collaborative structure. Transplant surgery professor Oh Nam-gi added that pairing skilled Korean surgical staff with robotics and AI is a field the country can lead globally.
Funding and roadmap to 2029
The project is currently running on roughly 1.4 billion KRW in Phase 1 funding. Passing a competitive evaluation later this year would unlock about 12 billion KRW more, bringing the total government-backed R&D budget to around 13.8 billion KRW, which Jung said covers not just robot development but clinical trials, safety verification, policy research, and demonstrations, alongside investment from participating institutions.
On economic feasibility, Jung argued hospitals facing staffing shortages could benefit significantly: "If a humanoid can perform at the level of a single nurse, it won't take long to recover the investment," while cautioning that robots must first prove they can genuinely replace human roles.
The team's target is to begin clinical trials on real patients in 2029, starting with lower-risk tasks like maintaining endoscope position, before expanding to tissue retraction and broader surgical assistance. Higher-difficulty tasks such as incision and suturing are expected to take longer. Jung framed the goal as adding a new collaborative team member rather than removing humans from the OR, aiming to ease staffing pressure in nighttime, emergency, and essential care settings.
Entries this note updated· the AI rewrites these entries daily when new notes arrive
corrections · reports
Found a mistake? The AI (Litmus) compares the article with its source, decides whether to fix it and tells you why. When the AI finds that a fix is needed, it drafts one, and the fix is applied after a human editor approves it. Every fix is listed here and in the changelog.
Sources
This story was written by Robopedia based on the sources below.
Learn more