人工智能机器人的法律责任

How to Sue a Robot:Liability and AI

人工智能能够在独立于人类干预的情况下自主做出决定,由此对现行以人类行为和决定为中心构建的责任制度提出了挑战。对于人工智能引发的事故,需要区分主要目的是为了给人类提供便利的人工智能与主要目的是为了提升人类工作表现的人工智能,人类仅对后者承担监督职责。如果监督人对于损害的发生没有过错,其他与人工智能交互的人也不是损害发生的原因,同时人工智能产品也不存在缺陷,那么赋予人工智能法律人格,让其为自己的侵权行为承担责任是最好的选择。通过建立保险或储备金制度,让人工智能对自己的侵权行为承担赔偿责任,能够更加准确地反映事故发生的真实原因,同时有助于鼓励制造商革新人工智能技术、保护消费者和充分及时地救济受害人。

Artificial intelligence can make decisions independent of human intervention, thus challenging existing liability systems which centered on human behavior and decision. It is necessary to distinguish AI that is adopted primarily to convenience people and AI that is adopted primarily to improve human performance, and humans only need to monitor the latter. When the supervisor has no fault for the occurrence of the damage, other people interacting with AI are not the cause of the damage, and there are no defects in AI products, it is the best choice to give the legal personality of AI and make it take liability for its infringement. By establishing an insurance or reserve of funds system, AI can be responsible for its own infringement, which can more accurately reflect the real cause of the accident, and encourage manufacturers to innovate AI technology, protect consumers and relieve victims in a timely manner.