摘要
人工智能对现行民法存在多维度的挑战,但尚不足以突破其边界。人工智能技术种类较为宽泛,需予以类型化,应将人工智能基础技术排除在法学讨论范畴外;人工智能无法享有主体资格和精神利益,类比动物、法人而承认人工智能的民事主体地位的路径难以实现理论的自洽。人工智能生成物能否认定为作品需根据输出内容是否具有高度同质性分类讨论,在能认定为作品的情形下,著作权人为人工智能保有人。"类人型"人工智能在缔约、履约过程中可定位为"传达人"。自动驾驶汽车与一般机动车在产品责任与机动车交通事故责任认定上并无本质区别。人工智能本身并不符合高度危险责任的标准,但当其载体为高度危险源时,高度危险责任有适用空间。
Artificial intelligence has multiple dimensions to the current civil law,but it is not enough to break through its boundaries. The types of artificial intelligence technology are relatively broad and need to be typed. The artificial intelligence basic technology should be excluded from the scope of legal discussion;artificial intelligence cannot enjoy the subject qualification and spiritual interests,and the path of recognizing the civil subject status of artificial intelligence is analogous to animals and legal persons. Realize the self-consistency of theory. Whether the artificial intelligence product can be identified as a work needs to be discussed according to whether the output content has a high degree of homogeneity. In the case that it can be identified as a work,the copyright owner is an artificial intelligence person. "Human-like" artificial intelligence can be positioned as "communicating person" in the process of contracting and fulfilling the contract. There is no essential difference between autonomous vehicles and general motor vehicles in terms of product liability and motor vehicle traffic accident liability. Artificial intelligence itself does not meet the highrisk liability standard,but when its carrier is a high-risk source,high-risk liability applies space.
作者
冉克平
谭佐财
Ran Keping;Tan Zuocai
出处
《私法》
2019年第2期352-370,共19页
Private Law Review
关键词
人工智能
民事行为
类型
效果
Artificial Intelligence
Civil Behavior
Type
Effect
