生成式人工智能网络小说的性质认定和权属论证

The Nature Recognition and Ownership Argumentation of Generative Artificial Intelligence Network Novels

摘要: 人工智能在出行、就医、娱乐等方面潜移默化的影响着人们的生活，以ChatGPT为代表的生成式人工智能也为网络小说写作提供了便利。随之而来的是生成式人工智能技术发展与现有著作权法律制度之间的冲突，其中最重要的著作权问题有两个：生成式人工智能创作的网络小说是否属于著作权法上所称的作品？该作品的权利和义务又归属于哪一主体？这两个问题富有争议且紧密相关，是研究其他问题的基础，文章从生成式人工智能运行过程中多主体的利益平衡角度入手，结合司法案例分析生成式人工智能的法律地位，从公平与效率的角度确定分配状态，按照主体对生成式人工智能作品的实质智力劳动比例来分配著作权归属。文章对文本生成过程中的数据挖掘和数据分析涉及对其他已有作品的著作权侵权问题提供解决思路，从法律的层面促使生成式人工智能的利用和可持续发展。

Abstract: Artificial intelligence has a subtle impact on people’s lives in areas such as travel, medical treatment, and entertainment. Generative artificial intelligence, represented by ChatGPT, has also provided convenience for online novel writing. Subsequently, there is a conflict between the development of generative artificial intelligence technology and the existing copyright legal system. Among them, the most important copyright issues are two: whether the online novels created by generative artificial intelligence belong to the works referred to in copyright law? Which subject does the rights and obligations of this work belong to? These two issues are controversial and closely related, serving as the basis for studying other issues. This article starts from the perspective of balancing the interests of multiple parties in the operation of generative artificial intelligence, and analyzes the legal status of generative artificial intelligence through judicial cases. From the perspectives of fairness and efficiency, the distribution status is determined, and copyright ownership is allocated based on the proportion of substantive intellectual labor of the parties to the work of generative artificial intelligence. This article provides solutions to copyright infringement issues related to data mining and analysis in the text generation process, promoting the utilization and sustainable development of generative artificial intelligence from a legal perspective.