In the discussion of ethics, armed drones and post-heroic war thus far, the focus has been on human decisions and actions. The availability of technology for remote-control killing presents opportunities for political leaders to convince themselves and their citizens that resorting to armed force carries no risk, and in the physical separation of weapon and warrior there is scope both to enhance and to degrade adherence to principles governing the conduct of war. In this chapter, the discussion turns to an as-yet theoretical circumstance in which the weapon is the warrior. In the United States, there is strong technological and institutional momentum towards the development and use of autonomous drones, but there is also deep concern there and elsewhere over the prospect of machines that can decide to kill humans. From an ethical perspective, an argument in favour of autonomy might be that, given the poor record of human adherence to just war principles, an armed drone could be programmed to do a better job. An alternative would be to hold fast to the notion that both war and ethics are necessarily and inescapably a human affair. Thus, when contemplating autonomous drones as a seemingly ‘post-human’ approach to war, the critical issue is whether or how technology can overcome ethical shortcomings in the use of force while preserving the moral influence of human responsibility. In government docu-ments and academic literature on military robotics, little differentiation is made between machines that operate on land, in water or in the air. The latter element, however, is arguably worthy of special attention when considering the post-heroic appeal of low-risk uses of force. Not only does an airborne robot relieve a human pilot of a potentially dangerous job, it is itself less vulnerable to the attacks of earth-bound adversaries. Autonomous drones — having greater capabilities for speed, penetration, perception and manoeuvre — are likely therefore to be preferred over land-based robots in much the same way that traditional aircraft are currently often preferred over ground troops. The making of this distinction on a strategic or operational basis does not, however, preclude ethical consideration of autonomy in robots more generally. What matters most for present purposes is the capacity of any machine, airborne or not, to decide to kill.