以DeepSeek为例论人工智能生成内容的著作权否定

The DeepSeek as an Example of Copyright Negation for AI-Generated Content

摘要: 近日，DeepSeek模型的横空出世再度引起人工智能生成内容可著作权性的广泛讨论。就目前所代表的生成式人工智能的运作过程及其显著特征而言，其输出信息无疑暴露出高度自动化和算法驱动的本质。确定人工智能生成物的著作权定位需要对现行法与比较法进行。分析反思。生成式人工智能不具备作者的法律资格，且其内容因不具备独创性而不具有著作权法意义上的作品属性。尽管人工智能生成内容在著作权法框架下可能难以直接获得保护，但通过不正当竞争法、数据财产保护等法律渠道，其创造者、所有者和使用者仍可获得一定程度的法律保护。在促进知识共享和创新的时代背景下，将人工智能生成内容视为开放知识资源留在公有领域，能够促进知识传播与交流，推动信息时代人类文化、艺术和科学的繁荣。

Abstract: Recently, the emergence of the DeepSeek has once again sparked widespread discussions on the copyright nature of AI-generated content. In terms of the operation process and salient features represented by current generative artificial intelligence, the output information undoubtedly reveals its highly automated and algorithm-driven nature. To determine the copyright positioning of AI-generated content, it is necessary to conduct analysis and reflection on current laws and comparative laws. Generative artificial intelligence does not have the legal status of an author, and its content lacks the work attributes in the sense of copyright law due to the lack of originality. Although AI-generated content may be difficult to obtain direct protection under the copyright law framework, through legal channels such as unfair competition law and data property protection, the creators, owners, and users can still obtain a certain degree of legal protection. Against the backdrop of promoting knowledge sharing and innovation, regarding AI-generated content as open knowledge resources in the public domain can promote the dissemination and exchange of knowledge, and drive the prosperity of human culture, art, and science in the information age.