Causal Liability in Autonomous Systems
摘要
With the widespread use of autonomous systems, liability often shifts towards the manufacturers and part suppliers, especially when various system components, sourced from different suppliers, contribute to a failure. This necessitates a framework for automatic liability apportionment which can be embedded in manufacturer-supplier contracts and minimise legal disputes. To this end, we propose a formal framework based on the notion of actual causality in structural causal models and the robustness semantics of logical specifications. We prove several desirable properties of this framework. Moreover, we formalise the notion of causal non-interaction, give sufficient conditions for it to hold, and demonstrate its utility in deriving upper bounds for liability analysis. Furthermore, we relate our definition to the existing notion of harm, empirically evaluate our framework to demonstrate its efficacy, and release a software package implementing our approach. We extend our framework to handle interval specifications.