Humanoid Robots Are Operating on People Now

A humanoid robot just removed someone's gallbladder. Not a specialized surgical arm. Not a da Vinci system with limited mobility. An actual humanoid—two of them, in fact—working in tandem during a preclinical trial at UC San Diego.

The results, published in Nature, represent something more significant than another incremental advancement in robotic surgery. For decades, surgical robotics has meant purpose-built machines: systems designed exclusively for operating rooms, with specialized end effectors and limited range of motion. The da Vinci system, despite its prevalence in over 7,000 hospitals worldwide, can't do much beyond what it was explicitly programmed to do. It's a tool, not an agent.

Humanoids change that equation entirely. A robot with human-like form and dexterity doesn't need a redesigned operating theater. It doesn't require specialized instruments. It can theoretically use the same tools, navigate the same spaces, and adapt to the same unpredictable scenarios that human surgeons face. In the UC San Diego trial, one humanoid assisted a human surgeon while two others operated autonomously—a demonstration of flexibility that specialized surgical systems simply cannot match.

But here's what the researchers aren't saying out loud: if humanoids can do surgery, they can do everything else in a hospital too. The same robot that removes your gallbladder could theoretically prep the operating room, transport patients, manage post-op care, or assist with rehabilitation. The form factor isn't just about surgical capability—it's about universal utility in human-designed environments.

The immediate reaction will focus on safety, and rightly so. Teleoperated surgery introduces latency concerns. Autonomous surgical decisions raise liability questions that make autonomous vehicle debates look simple. Who's responsible when a humanoid makes a clinical error? The hospital? The manufacturer? The supervising physician who may not have even been in the room?

Yet these questions miss the broader trajectory. Specialized surgical robots took decades to gain acceptance because they required hospitals to redesign workflows, retrain staff, and invest in single-purpose infrastructure. Humanoids bypass all of that. They slot into existing systems. They use existing tools. They work alongside existing staff. The barrier to adoption isn't technical capability anymore—it's regulatory approval and institutional trust.

The researchers note their humanoids performed the gallbladder removal successfully, but they're careful to position this as augmentation rather than replacement. That framing won't last. Once humanoids prove they can match or exceed human surgical outcomes—and given the pace of AI development, that's a when, not an if—the conversation will shift from "Can robots assist surgeons?" to "Why do we need human surgeons for routine procedures?"

We've seen this pattern before in radiology, pathology, and diagnostic medicine. AI starts as a helper. Then it becomes better than humans at specific tasks. Then it becomes the default, with human oversight positioned as the safety net. Surgery always felt like it would be different—too complex, too improvisational, too dependent on human judgment. The UC San Diego trial suggests that assumption was wrong.

The real story isn't that robots can do surgery. It's that they can do it using our tools, in our spaces, following our protocols. That's not a feature. That's a paradigm shift. And it's happening faster than anyone in the medical establishment is ready to admit.