A clinician can operate a robot from another room while still controlling its movements. A robot can also perform a defined task with some autonomy while a clinician remains nearby. FDA’s November discussion paper treats autonomy and remote teleoperation as different dimensions, not interchangeable descriptions of “robotic surgery.” Discussion paper, pages 3–7

That distinction changes the question an explanation needs to answer: who is performing the task, where are they, and how does control move when circumstances change?

Control and distance describe different things

Control and distance describe different things
DimensionWhat the paper is discussingA separate question it leaves open
AutonomyA device performing tasks or making decisions from its current state and sensing, without user interventionWhich tasks, adaptations and decisions are transferred?
Remote teleoperationA user controlling the system in real time while outside the patient’s roomWhat happens when a local intervention is needed?
Both togetherAutonomous functions used within a remotely operated systemHow are delays and handoffs between modes handled?

These are explanatory categories from one FDA discussion, not a rating of individual systems. A commercial product name or the presence of a console does not, by itself, establish its level of autonomy.

The paper also distinguishes autonomy from automatic behavior that follows established rules. It discusses several frameworks rather than adopting one universal scale as a new regulatory requirement. Terminology and autonomy, pages 6–8

Moving a task is not the same as moving expertise

For remote operation, the paper asks about connection loss, delayed signals, communication and sensory information unavailable to the distant operator. It also asks who can intervene locally if a procedure must change or stop.

That creates a practical tension in the proposed benefit: remote expertise may expand access, but requiring another specialist to remain on standby could consume some of the capacity being shared. FDA raises that as a benefit-risk question, not a measured conclusion about every remote service. Remote teleoperation, pages 11–14

For autonomy, the emphasis shifts to the tasks the system performs, its response to unexpected conditions and the user’s ability to supervise or interrupt it. Distance may be unchanged even while decision-making authority changes.

A handoff can join the two problems

The paper considers additional risks when remote operation and autonomy are used together. An autonomous action may need human intervention at a moment when communication is delayed; switching modes can introduce its own failure possibilities.

The useful comparison is therefore more specific than asking whether a robot is “in control.” It identifies the task, the user’s role, the connection and the point at which another person or mode must take over. This is a way to describe the evidence question, not a checklist establishing that a particular system is safe to deploy.

Our clinical-AI authority analysis addresses the related distinction between a capability and permission to act. Physical robotic systems add questions about instruments, anatomy, sensing and immediate intervention.

The document is a discussion, not a new permission

FDA explicitly states that the paper is not draft or final guidance, does not propose or implement policy changes and does not communicate new regulatory expectations. The agency requests feedback by 5 January 2027. Status and feedback

That makes it different from the annual guidance-priorities list and from a device-specific marketing decision. A discussion of future capabilities does not show that those capabilities are authorized, available or effective in a named hospital.