A humanoid robot performed surgical tasks on live human tissue, guided by a surgeon watching from elsewhere.
In December 2024, researchers at the University of California watched a robot with dexterous hands and cameras for eyes work through a segment of pig intestine. The robot sutured tissue while a surgeon sat remote, guiding movements through a control interface.
This is not the robot thinking through anatomy—this is not the robot reading the tissue, understanding damage, deciding what suture pattern will hold. The surgeon is doing that. The robot is the surgeon's hands extended through space and time delay, with better precision than human hands can achieve.
The surgeon watching the remote feed sees something fundamentally different from standing over the patient. Edmund de Waal, the ceramicist, has described how his hands know things his conscious mind does not—when you throw clay, you feel resistance that your eyes cannot capture. When you stand over tissue, your body reads pressure and give and texture in ways no camera transmits back. The remote surgeon gets data.
When the surgeon is remote and the robot fails, the question becomes slippery.
This is why the protocol matters. When surgeons work directly, they are responsible—they stand in the consequence of what their hands created. When the surgeon is remote and the robot fails, the question becomes slippery. Did the surgeon misread the interface? Did the robot misexecute?
The robot here is not revolutionary. It is clarifying. It shows that we have learned to separate the person who decides from the person whose body carries weight. The question for anyone building a practice is whether your system does the same. Whether you have split decision-making from consequence. Whether that split is actually removing error or just removing your ability to feel when something goes wrong.