Much of the research and proposed industrial deployment of Remote Operations (ROs) in support of automated vehicles is founded on the optimistic premise that in-vehicle standby drivers and Safety Officers (SOs) can easily be replaced with ROs, with some commercial models proposing that a single RO supervise over 30 vehicles. However, emerging evidence suggests that the RO task is fundamentally different from the in-vehicle driving task. Furthermore, communications latency and reliability constraints, coupled with fragmented attention and altered task demands, introduce distinctive human factor challenges. These include degraded situational awareness, increased cognitive workload, and reduced capacity for timely intervention. The result is a widening gap between what is commercially desirable and what may be operationally appropriate. This paper argues that the central question for remote operation in support of automated vehicles is not one of technical feasibility but of human-centred appropriateness, and debates which RO roles should continue to be developed and which should be constrained or avoided. We present a synthesis of research on remote vehicle operations, identifying recurring human-factor limitations and mapping them to proposed remote tasks. The paper concludes with targeted recommendations for designers, operators, and regulators intended to question the scaling of teleoperation models and to reframe the debate from “Can we teleoperate?” to “Under what conditions should we?”