论人形机器人治理中的刑法归责

On the Attribution of Criminal Liability in the Governance of Humanoid Robots

随着机器人和人工智能技术的快速发展,作为具身智能的人形机器人在应用场景上逐渐拓宽,同时也形成了多元而复杂的安全风险和刑法归责难题。理论上主要存在代理责任、过失责任、严格责任、独立责任四种各有利弊的刑法归责模式。除了严格责任之外,其余模式都可以统合在场景化的刑法归责体系之中。在当前人工智能的发展阶段,传统的刑法教义学原理可以处理绝大部分的机器人刑法归责问题,不过也应在新的技术条件下对可容许风险、信赖原则予以新的推演和发展。取决于未来的技术进展,在功能主义的视角下,将来可以考虑有条件地肯定智能人形机器人独立责任主体地位的可能性。人形机器人的刑法归责体系,应当对其他法秩序的规范评价和机器人伦理标准的探讨保持协同和开放。

With the rapid development of robotics and Artificial Intelligence technology,humanoid robots,as embodied intelligence,are gradually expanding their application scenarios,leading to diverse and complex safety risks and challenges in the attribution of criminal liability.Theoretically,there are four main models of criminal liability:agency liability,negligence liabili ty,strict liability,and independent liability,each with its advantages and disadvantages.Except for strict liability,the other models can be integrated into a scenario-based criminal liability system.At the current stage of AI development,traditional doctrinal principles of criminal law can handle the majority of issues related to criminal liability for robots.However,there should also be new interpretations and developments based on permissible risks and the principle of reliance under new technological conditions.Depending on future technological advancements,under a functionalist perspective,there may be possibilities to conditionally recognize the independent liability status of intelligent humanoid robots.The criminal liability system for humanoid robots should be concurrent with and open to the normative evaluation of other legal systems and the exploration of robot ethical standards.