Robot Surgeons Tackle Live Organs

The first live gallbladder surgeries performed in pigs by teleoperated humanoid robots are less about a flashy “world first” and more about a quiet but consequential redesign of how we might bring advanced surgery into ordinary, cramped, and underserved spaces.

Key Points

UC San Diego’s team teleoperated humanoid robots to complete two laparoscopic gallbladder removals in live pigs, demonstrating in vivo feasibility rather than mere lab simulation.
The humanoid robots, roughly human-sized, work in standard operating rooms and tight environments where existing bulky robotic platforms struggle, opening a path to wider surgical access.
One procedure paired a humanoid with a human assistant; the second used two humanoids together, showing both human–robot and robot–robot team configurations.
The work is an early-stage, preclinical proof of concept with only two cases; it is not yet validated or approved for human patients and faces significant regulatory and technical hurdles.
This milestone sits within a broader evolution of surgical robotics, including fully autonomous systems like Johns Hopkins’ SRT-H, and will be shaped by stringent frameworks for evaluating innovation in the operating room.

From console robots to humanoids: what really changed?

Robotic surgery is not new; what is genuinely novel here is the combination of humanoid form factor with live, in vivo surgery. For two laparoscopic cholecystectomies (gallbladder removals) in pigs, surgeons at UC San Diego controlled general-purpose humanoid robots to perform the full procedure inside a real operating room. This is a qualitative step beyond benchtop or cadaver experiments: the robots had to contend with living tissue, real physiology, and the practical choreography of an OR.

Unlike platforms such as the da Vinci system, which are purpose-built surgical consoles occupying substantial floor space, these humanoid robots are roughly five feet tall, with human-like arms that hold standard laparoscopic instruments. They stand where a human assistant would stand and reach into the same narrow operative field, instead of requiring the room to be rebuilt around them. That is the core insight: rather than redesigning the hospital for the robot, the robot is redesigned for the hospital.

How the humanoid surgical system actually works

The UC San Diego system is teleoperated, meaning a human surgeon remains firmly “in the loop.” The robot’s arms and instruments are driven by a surgeon at a console—typically wearing a VR headset or using joysticks—while the robot translates those motions into precise movements in the patient. This is conceptually similar to existing surgical robots, but the physical embodiment is different: the actuators and joints are packed into a human-sized frame, not a multi-ton base and boom.

In the reported preclinical trial, two configurations were tested. In one, a humanoid robot performed the laparoscopic cholecystectomy with a human bedside assistant who physically adjusted the robot’s arms and helped manage instruments. In the second, a pair of humanoid robots worked side by side, eliminating the human assistant role and demonstrating a “robot–robot team” surgery. Endoscopic video shows the robots dissecting tissue around the gallbladder and liver, placing clips, and cauterizing cystic structures with the sort of fine, stable motion expected of modern teleoperated systems.

Importantly, the technical paper and accompanying press material describe this work as a feasibility study. The authors emphasize that performance is comparable in accuracy to gold-standard research platforms like the da Vinci Research Kit (dVRK), but with slower execution and higher operator demand, largely due to the complexity of controlling humanoid hardware. In other words, the system can do the job, but not yet with the efficiency or ergonomic polish of mature surgical robots.

Why humanoid form factor matters for surgical access

The choice to build a humanoid robot, rather than another specialized surgical console, is not about mimicking people for its own sake. It is about deployability. A humanoid with human-scale arms can step into the operating room layouts that already exist in small hospitals, rural clinics, and even ships, using the same carts, the same instrument trays, the same bed positioning. That matters because most of the world does not have the space, money, or infrastructure for a dedicated robotic suite.

Conventional platforms are large, expensive, and logistically demanding. Installing them can require structural modifications, long staff training, and changes to patient flow. By contrast, a humanoid assistant could be rolled into a standard OR as an incremental addition—first handing instruments and suctioning, later participating in actual tissue dissection. The UC San Diego team explicitly links this design to the goal of extending advanced surgery to isolated communities where space and skilled personnel are limited, and where remote operation from a distant center could be transformative.

In that sense, the humanoid robot is less a rival to current robots than a new category: a general-purpose clinical worker that can both manipulate instruments in the sterile field and, in principle, perform selected parts of procedures under supervision. It supports a vision of “surgical telepresence” in which expertise is centralized but hands are distributed.

Autonomy versus teleoperation: placing this work in the wider robotics landscape

Humanoid gallbladder surgery does not happen in a vacuum. In parallel, other groups have pushed toward full autonomy. Johns Hopkins University’s SRT-H system, for example, has performed the clipping and cutting portion of a laparoscopic cholecystectomy autonomously on pig gallbladders, achieving 100% success across eight cadaveric trials. The robot, trained on videos of human surgeons and guided by a hierarchical AI policy, responds to voice commands, adapts to anatomical variation, and self-corrects when conditions change.

This contrast is instructive. The UC San Diego humanoids are teleoperated and live-tissue tested, focusing on feasibility in real OR conditions. SRT-H is more autonomous but remains in a lab-like environment, working on pig organs and lifelike models without full in vivo constraints. Together, they outline a spectrum: from specialized autonomous task-robots to versatile teleoperated humanoids. The most likely future is neither pure autonomy nor pure teleoperation, but a hybrid in which humanoids execute low-level motions and workflows semi-autonomously, while surgeons retain high-level control.

For clinicians, the bottom line is capability and reliability, not the robot’s appearance. Yet the humanoid form factor can simplify integration: if a system occupies the footprint of a human assistant and uses familiar tools, resistance from OR staff is lower than for a massive new console, especially in resource-constrained settings.

Surgeons at UC San Diego just handed the scalpel to two humanoid robots, who went on to complete live surgical procedures for the first time in histo…https://t.co/bPVDg0orbV pic.twitter.com/UslGzNE898

— New Atlas (@nwtls) July 10, 2026

Evidence strength and the limits of a two-surgery trial

The core scientific claims around the UC San Diego study are unusually well supported and, so far, uncontested. A peer-reviewed paper in Nature describes the in vivo porcine cholecystectomies and explicitly labels them as the first humanoid-based in vivo surgical trial of this type. The arXiv version of the study outlines bench experiments, dry-lab work, and the progression into animal surgery, detailing performance metrics across platforms. University press releases and media coverage converge on the same key facts: two surgeries, one human–robot team and one robot–robot team, both completed successfully.

At the same time, the study’s limitations are plain. A sample size of two animals cannot establish complication rates, long-term outcomes, or robust safety margins. No adverse events are reported, but absence of evidence at this scale is not evidence of absence in a real-world population. Moreover, at least one configuration required a human bedside assistant to reposition robot arms, highlighting that this is not yet a drop-in replacement for human staff.

There is, at present, no primary-source counter-evidence challenging the fact of these surgeries or the basic performance claims. No rival group has published a forensic critique of the surgical videos, no regulatory body has raised formal concerns about misrepresentation, and no technical refutation appears in the literature. That makes the “world first” label for humanoid in vivo surgery reasonably secure within its narrow scope. The necessary skepticism should attach not to whether the surgeries happened, but to how far those two cases can be extrapolated.

Regulation, IDEAL, and the long road to human trials

Moving from pigs to people is not a matter of media enthusiasm; it is a matter of regulatory evidence. Contemporary surgical innovation is often evaluated through frameworks like IDEAL (Idea, Development, Exploration, Assessment, Long-term study), which lay out staged requirements for devices and techniques. Under such frameworks, preclinical animal work is an early step, followed by pilot human studies, controlled comparative trials, and long-term outcome tracking.

Most experimental robotic platforms never complete that journey. Analyses of surgical device development suggest that fewer than roughly 15% of novel systems that reach porcine trials ultimately obtain FDA approval for routine human use, reflecting failures in efficacy, safety, economics, or workflow fit. Humanoid teleoperated robots will have to satisfy not just technical criteria but stringent human factors assessments: can surgeons work with them without excessive cognitive load, do they fit into existing staffing patterns, and can hospitals justify the cost relative to conventional robots or manual surgery?

Regulators will also scrutinize remote operation proposals. Latency, telemetry security, failover when networks drop, and liability for harm when surgeon and patient are separated by hundreds of miles are nontrivial hurdles. The UC San Diego team has expressed interest in remote deployment to isolated communities, but real data on latency tolerance and clinical performance under remote conditions have yet to be published. Those numbers will matter more to approval bodies than any “historic” headline.

Media narratives, public perception, and naming confusion

Coverage of the humanoid surgeries has oscillated between sober explanation and sensationalism. University communications and many science outlets emphasize the preclinical, proof-of-concept nature of the work. Popular media and social posts, by contrast, lean heavily on “world first” and “historic” framing, sometimes implying a readiness for human trials that the researchers themselves explicitly disavow.

Even the robot’s name reflects this gap. Reports and broadcasts variously refer to the system as “Surge,” “Sergie,” or “Surgie,” reflecting inconsistent branding in public channels. This matters less for the science than for traceability: when investors, regulators, or clinicians try to follow the technology’s development, fragmented naming can blur which prototype or configuration is under discussion, and social media exaggeration can inflate expectations beyond what the data support.

Researchers have tried to counter the “backflips and kung fu” stereotype that haunts humanoid robotics, stressing that this work is about practical societal utility rather than spectacle. For a skeptical public—especially those who have seen prior hype cycles come and go—the most convincing arguments will be not press releases but independent replications, comparative trials against existing systems, and clear demonstrations of improved access or outcomes.

What this means for surgeons, hospitals, and patients

For clinicians, the immediate impact of humanoid gallbladder surgery in pigs is modest but real. It expands the design space for future surgical robots and signals that human-scale, teleoperated humanoids can perform delicate laparoscopic maneuvers reliably in vivo, not just on models. For hospital administrators, it raises the prospect of robotic assistance that does not require a new building or a six-figure capital expenditure for a monolithic console.

For patients—especially those far from tertiary centers—the implications are more long-term. If humanoid platforms mature, they could underpin networks of remote surgical care, where a specialist at a regional hub operates a humanoid assistant in a distant clinic, much as teleradiology and teleICU services already extend expertise across geography. That will demand careful attention to equity: ensuring that “remote surgery” is not a lower standard of care for the poor, but a way of delivering the same standard to those currently excluded.

The real measure of success for this line of research will not be the next headline but the answer to a simple question a decade from now: did humanoid surgical robots become routine tools that quietly expand access and improve outcomes, or did they join the long list of impressive prototypes that never cleared the regulatory, economic, and human-factors bar for everyday medicine? The two pig gallbladders are an important start; they are not yet the finish line.