摘要
以ChatGPT为代表的生成式人工智能自诞生起便对传统著作权法的制度规则发起挑战,而首当其冲需要明确的是人工智能生成内容在著作权法中的定位问题。人工智能生成内容不仅包含人类用户的独创性贡献,人工智能本身也作为内容产出者参与输出表达。从其本质来看,人工智能生成内容具备表达外观,是满足独创性要求的智力成果,具备可版权性。实践中应当破除多模态生成式人工智能的工具论思想,结合产业实际,灵活判断涉案人工智能生成内容在著作权法中的定位,引导生成式人工智能可靠可控发展。
The generative AI represented by ChatGPT has challenged the traditional rules of copyright law since its birth,and the first thing that needs to be clarified is the positioning of AI generated content in copyright law.AI-generated content not only contains the original contributions of human users,but the AI itself also participates in the output expression as a content producer.By its very nature,AI-generated content has the appearance of expression,which is an intellectual achievement that meets the requirement of originality and is copyrightable.In practice,the instrumentalist idea of multimodal generative AI should be cleared,and the position of the AIgenerated content in the copyright law should be flexibly judged together with the industrial reality,so as to guide the reliable and controllable development of generative AI.
作者
冯晓青
李可
Feng Xiaoqing;Li Ke
出处
《中国版权》
2024年第3期20-30,共11页
China Copyright
基金
北京市习近平新时代中国特色社会主义思想研究中心重点项目“习近平法治思想中知识产权保护重要论述研究:理论体系与实践应用”(项目编号:22LLFXB038-2022)的阶段性成果。
