Humanoid robots controlled by surgeons did live-pig surgeries

Humanoid robots teleoperated by surgeons removed gallbladders from live pigs in a preclinical trial published in the journal Nature, completing two minimally invasive surgeries. The experiment used Unitree G1 humanoid robots, nicknamed "Surgie," with surgeons controlling the robots from a PC console with a stereo headset and a foot pedal.

What exactly did the surgeons and robots do?

The team completed two minimally invasive cholecystectomies on live pigs, using teleoperated Unitree G1 humanoid robots adapted to hold surgical instruments and controlled from a surgeon console. The first operation included a human surgeon standing alongside the robot as an assistant, and the second featured two teleoperated robots working together.

Surgeons used software that mapped intuitive hand motions to the robot wrists, and physical adapters were built so the robots could grasp standard surgical tools. The control station provided a stereo display, and a foot pedal engaged or disengaged surgeon hand motions from the tool movement.

How do these humanoid systems compare on cost, size, and regulation?

Unitree G1 humanoid robots start at a baseline price of $13,500, with shipping costs between $300 and $1,200, while upgrades such as dexterous hands can push costs beyond $67,000, making them substantially cheaper than specialized surgical systems. By contrast, Intuitive Surgical’s da Vinci Surgical System can cost between half a million dollars and several million dollars.

The Unitree G1 stands about 5 feet tall and weighs roughly 60 pounds, with an arm span of 450 millimeters. Specialized surgical robots can weigh about 1,800 pounds and occupy significantly more space in operating rooms. The da Vinci system has been cleared by the US Food and Drug Administration and other medical regulatory agencies and has been tested in multiple clinical trials; the teleoperated humanoid approach remains experimental and preclinical.

What technical limitations did the trial reveal?

The trial required frequent pauses for recalibration and for physically repositioning the robot bodies or arms, which made the surgeries take much longer than when performed with existing specialized surgical systems. The Unitree G1’s compact arm span and constrained range of motion limited reach for remote operators and increased both cognitive and operational workload for the surgical team.

Current teleoperated humanoid systems typically exhibit latencies in the hundreds of milliseconds, while the researchers note previous studies suggest surgical robots should ideally have latency below 150 milliseconds. Trainees and experienced surgeons performed practice tasks faster on da Vinci Research Kit hardware compared to controlling the humanoid robots, underscoring current performance gaps.

Why it matters

Lower-cost, smaller-format humanoid robots could expand access to robotic-assisted surgery in settings that cannot afford or do not have room for specialized systems, from rural hospitals to battlefield or space deployments. As Shanglei Liu, an assistant professor of surgery at UC San Diego, said, "It’s a fraction of the cost and it takes a fraction of the space in an operating room." If the technical limits of reach, latency, and reliability are addressed, teleoperated humanoids could let expert surgeons extend care to underserved locations.

At the same time, the da Vinci platform remains a regulatory and performance benchmark, already cleared by agencies and proven in clinical trials. The humanoid approach must close gaps in latency, ergonomics, and task speed before it can be considered for human patients or regulatory review.

What to watch

Track improvements that reduce pauses and recalibration, upgrades that add dexterous hands and longer reach, and efforts to lower control latency toward or below 150 milliseconds. A move from preclinical animal trials to regulated human trials, or demonstrations showing parity with da Vinci Research Kit task speeds, would be the next concrete milestones to judge clinical readiness.