by Fan Yang
Abstract: The militarization of Lethal Autonomous Weapons Systems inevitably triggers “agency spillover engagements”, situations where algorithmic agency deviates from human presets. This has led to systemic applicability ruptures across International Humanitarian Law, state responsibility, and International Criminal Law. This paper argues that the root cause of the failure lies in the strict adherence to the human moral agent presumption. To address this, this paper proposes an ontological perspective of human machine hybrid agency and employs the principal-agent relationship as an analytical lens. It further derives three domain-specific tools – the Joint Cognitive System, Risk Allocation, and Reckless Delegation – to transcend these dilemmas within lex lata through evolutionary and teleological interpretation.
Keywords: Lethal Autonomous Weapons Systems; Agency Spillover Engagement; Hybrid Agency; Principal-Agent; International Humanitarian Law; Legal Interpretation
