top of page

The Surgical Autonomy Ladder: What Levels 0 Through 5 Actually Mean

  • Aug 12
  • 3 min read

Updated: 4 days ago

The Surgical Autonomy Ladder: What Levels 0 Through 5 Actually Mean

Ask ten surgeons what “autonomous surgery” means and you may get ten different answers.


For some, it means a robot operating without a surgeon. For others, it means a robot independently completing a single surgical task.


Those are very different capabilities.


A six-level framework, from Level 0 to Level 5, provides a way to distinguish them. The important variable is not how advanced the robot looks. It is who makes the decisions and who executes the surgical actions.


Level 0: Manual


The surgeon makes all decisions and controls all movements.


The robot, if present, functions as an instrument. It provides no autonomous decision-making or execution.


Real example:


AESOP, built by Computer Motion, was cleared by the FDA in 1994 as a voice- and foot-pedal-controlled arm that held the laparoscope. It moved the camera exactly where the surgeon told it to go.



Level 1: Assisted


The system modifies or improves the surgeon's movements.


Examples include motion scaling, tremor filtering, or other forms of robotic assistance.


The surgeon remains responsible for every action.


This is where most clinical robotic surgery currently sits.


Real example:


The da Vinci Surgical System, from Intuitive Surgical, filters the surgeon's hand tremor and scales large hand movements into small instrument movements.


Da Vinci is the clearest real-world proof of where most clinical robotic surgery actually sits.



Level 2: Shared Control


The system performs a defined function while the surgeon remains actively involved.


The robot may execute a constrained subtask, but the surgeon continuously directs the procedure and can intervene.


This is the transition from assistance to limited autonomy.


Real example:


Stryker's Mako SmartRobotics uses AccuStop haptic technology to enforce a pre-planned virtual boundary during knee and hip replacement.


The surgeon still moves the saw or burr by hand. The system physically resists the instrument if it drifts outside the plan.


Source: Stryker


Level 3: Supervised Autonomy


The robot independently performs a defined surgical task.


The surgeon supervises the system and remains responsible for intervening when necessary.


The autonomy is limited to a specific task, procedure, or set of conditions.


Real example:


The Smart Tissue Autonomous Robot, STAR, built at Johns Hopkins, completed more than 83% of the suturing task autonomously while reconnecting two sections of intestine in live pigs.


The system still required manual fine-tuning by the surgical team to correct positioning when it missed a stitch.


That mix, most of the task running independently with a human catching the misses, is what supervised autonomy looks like in practice.



Level 4: Conditional Autonomy


The robot performs larger portions of a procedure independently within defined conditions.


The surgeon moves from active control toward supervision and intervention.


No commercially available surgical robot has reached this level.


Real example:


A Johns Hopkins and Stanford team built SRT-H. This system autonomously performed the clipping-and-cutting phase of gallbladder removal. It responded to spoken corrections from the surgical team without anyone touching the controls.


This happened ex vivo, in a lab, not on a living patient, and no version of it is commercially available.



Level 5: Full Autonomy


The system plans and executes an entire surgical procedure without human supervision.


This remains a research objective rather than a clinical reality.


Real example:


None exists. Every system above still has a human either supervising, correcting, or standing by to intervene.


Where surgical robotics actually stands


The current landscape is considerably less autonomous than the term “autonomous surgery” can suggest.


Most commercially deployed surgical robots provide assistance rather than autonomous execution. Experimental systems have demonstrated higher levels of autonomy for specific surgical tasks, but that is very different from autonomous surgery at the procedure level.


That distinction matters.


A robot that autonomously completes one suturing task and a robot that independently plans and performs an entire operation belong to completely different categories.


The autonomy ladder makes that difference explicit.


The key question is how much decision-making and execution the robot actually controls.



For the full breakdown of where each cleared system sits on this ladder, see the Autonomy Registry. For a claim-by-claim check of what manufacturers say versus what their FDA filings support, see the Autonomy Claims Audit.



Related reading:




icon-autonomy.png

Autonomy

Four robots claim "autonomous." Here's where each one actually sits.

What's actually true about AI, autonomy, and surgical robotics?

Four reports. Every claim checked against FDA filings and peer-reviewed evidence — not press releases.

icon-ai-claims.png

AI

Five FDA-cleared systems. What's cleared, inferred, or just marketing.

The Evidence Library

icon-registry.png

Registry

Nine robots mapped across six levels of autonomy.

icon-roadmap.png

Roadmap

Intuitive's five-layer autonomy roadmap, audited layer by layer.

bottom of page