A combat drone is an unmanned aerial vehicle for combat purposes. It is a subclass of the combat robot. Its use ranges from fully remote-controllable long-range drones (human-in-the-loop), through semi-autonomous loitering munitions (human-on-the-loop), to fully autonomous systems with AI-supported target recognition (LAWS).
Generations (anonymized)
First generation (from 2001). Large long-range drones of US manufacture with satellite-supported remote control and precision-guided missiles. Used in particular in asymmetric conflicts from Afghanistan/Iraq onward. Cockpit and weapons release with the operator; autonomous navigation, no autonomous lethal decision.
Second generation (from approx. 2015). Medium combat drones of Turkish manufacture, deployed in several regional conflicts (Nagorno-Karabakh, Syria, Libya, Ukraine). More cost-effective, suited for mass production, still with human weapons release.
Third generation (from approx. 2020). Loitering munitions (kamikaze drones) with autonomous navigation, object recognition, and semi-autonomous target selection. Several models of US, Russian, and Turkish origin are known. As a rule human-on-the-loop; in borderline cases human-out-of-the-loop.
Fourth generation (from approx. 2023). AI-supported target image recognition, adaptive coordination algorithms, anti-jamming capability. In several cases (see LAWS) fully autonomous lethal use has been documented.
The most important deployment scenarios 2024–2026
- Ukraine war (since 2022). Massive deployment of various drone classes on both sides. Documented first autonomous AI lethal strike in the history of warfare in autumn 2023 (BBC, New Scientist). AI target recognition, swarm coordination, image classification.
- Gaza/Middle East. Deployment of AI-supported target identification systems with documented humanitarian consequences (The Guardian 2024).
- Libya 2020. UN panel-documented first public LAWS deployment against human beings (S/2021/229).
Assessment in terms of personhood ontology
As with the combat robot, the following holds: the drone itself is morally neutral — a tool. What is decisive is whether personal responsibility is realized or dissolved in its use.
- Human-in-the-loop: Attribution of responsibility intact. The operator decides, fires, answers for it.
- Human-on-the-loop: Attribution of responsibility burdened. The reaction time for intervention shrinks; the psychological readiness to escalate toward release increases (studies on automation bias, e.g. Parasuraman/Riley 1997).
- Fully autonomous: Attribution of responsibility breaks apart (responsibility gap after Sparrow 2007). Personhood ontology categorically rejects this mode, because it structurally institutionalizes oblivion of the person.
Ontological classification
Superordinate concepts: combat robot, robot
Typical degrees of autonomy: All four levels, depending on the model
Chapter assignment: Chapter 5: Oblivion of the Person
Sources: Generated by querying the Personhood ontology.
Further sources:
- Sparrow, Robert (2007): “Killer Robots.” Journal of Applied Philosophy 24(1), pp. 62–77.
- Parasuraman, Raja / Riley, Victor (1997): “Humans and Automation: Use, Misuse, Disuse, Abuse.” Human Factors 39(2), pp. 230–253.
- United Nations Panel of Experts on Libya (2021): Final Report S/2021/229.
- BBC News (2023): Reports on AI-supported autonomous drone strikes in the Ukraine war.
- New Scientist (2023): Report on the first documented autonomous AI lethal strike.
- Bode, Ingvild / Watts, Tom (2023): Loitering Munitions and Unpredictability. Center for War Studies Research Report.
- Chamayou, Grégoire (2015): A Theory of the Drone. New Press (philosophical foundational critique).
- The Guardian (2024): Reporting on AI targeting systems in Gaza.