西南交通大学学报（社会科学版）2026，Vol.27Issue(4)：15-24,148,11.
生成式人工智能服务平台承担版权过滤义务的正当性质疑
Questioning the Legitimacy of Copyright Filtering Obligations for Service Platforms of Generative AI
摘要
Abstract
In recent copyright infringement cases involving"Ultraman",the court ruled that the service platforms of generative AI should bear a copyright filtering obligation.Essentially,the service platforms of generative AI remain a form of service,and the obligations they should bear under copyright law should align with those of internet service providers(ISPs).Current legislation does not impose copyright filtering obligations on internet service providers.If the service platforms offer generative AI services specifically designed for copyright infringement,They can be directly regulated under the rules of contributory infringement in China's Civil Code.If the generative AI service is not specifically designed for copyright infringement,the"notice-and-takedown"rule still provides sufficient protection for rights holders,and there is no need to interpret the law as imposing a copyright filtering obligation on the service platforms of generative AI.Imposing such an obligation would place an excessive burden on these platforms,potentially hindering the development of generative AI technology and its industry.Therefore,the application of the law should not be interpreted as requiring the service platforms of generative AI to assume a copyright filtering obligation.关键词
生成式人工智能/网络服务平台/版权过滤义务/正当性Key words
generative artificial intelligence(generative AI)/generative AI service platforms/copyright filtering obligation/legitimacy引用本文复制引用
刘铁光,程骏宏..生成式人工智能服务平台承担版权过滤义务的正当性质疑[J].西南交通大学学报（社会科学版）,2026,27(4):15-24,148,11.基金项目
国家社科基金项目《知识产权民刑保护之法域冲突及其化解研究》(2021BFX197) （2021BFX197）
