How Should Autonomous Vehicles Allocate Accident Liability? Rethinking the Applicability of the Hand Formula

Abstract The Hand Formula internalizes tort externalities into decision-making by comparing prevention costs with accident damages, achieving theoretical breakthroughs in liability determination. However, in human-machine collaborative driving, AI’s “black box” nature and decision complexity make accident probabilities unpredictable and costs impossible to internalize, weakening institutional incentives due to extended decision chains and multiple liability subjects. To address this, negligence-based causal rules must evolve with changing care level concepts, while activity-based causal approaches better suit human–machine interactions. Institutionally, accident liability mechanisms should incorporate presumption of fault, establish multi-tiered liability allocation, and promote proactive insurance involvement. As AI develops, personalized rules will transform through intelligent evolution to become legal system fundamentals, while causation will serve both as a liability allocation tool and a mechanism for social stability.