Proefschrift

137 7 CONCLUSION Property 2: Human and AI agents have appropriate and mutually compatible representations of the human-AI system and its context. Property 3: The relevant agents have ability and authority to control the system so that humans can act upon their responsibility. Property 4: Actions of the AI agents are explicitly linked to actions of humans who are aware of their moral responsibility. In their reflection on their work the authors highlight that ‘Meaningful human control is necessary but not sufficient for ethical AI.’ (Cavalcante Siebert et al., 2022, p. 252). The authors amplify this by stating that for a human-AI system to align with societal values and norms, Meaningful Human Control must entail a larger set design objectives which can be achieved by transdisciplinary practices. All three approaches have in common that they focus on the human-machine interaction in order to operationalise the concept of Meaningful Human Control. Either by bridging the gap between weapon usage and ethical principles based on ‘if-then’ rules (Amoroso & Tamburrini, 2021), creating actional properties for the design of AI systems in which each of the properties human and artificial agents interact (Cavalcante Siebert et al., 2022), or proposing two LoA’s in which different agents have different levels of control over the decision-making process to deploy an Autonomous Weapon System (Umbrello, 2021). Mirroring these three approaches to the Comprehensive Human Oversight Framework we can conclude that they can be positioned in the socio-technical layer of the framework which describes the human-machine interaction. This entails that all three approaches disregard the governance layer of the Comprehensive Human Oversight Framework which describes the supervision processes for Human Oversight. Therefore, we can conclude that, although valuable for operationalising the concept for Meaningful Human Control, these approaches do not provide new insights for Human Oversight of Autonomous Weapon Systems in order to ensure accountability and responsibility. 7.2 LIMITATIONS The Comphrensive Human Oversight Framework introduced in this dissertation and the simulation of the implementation concept to operationalise the criteria for Human Oversight are new contributions of this research to the current academic work. Several limitations of these novelties can be identified. First of all, both the Comphrensive Human Oversight Framework as the simulation of the implementation concept are in this research only applied to one case: that of the deployment of an Aerial Autonomous Weapon Systems. The scenario and implementation concept are based on a tactical

RkJQdWJsaXNoZXIy MjY0ODMw