Copyright protection paradigm in the generative AI era: originality, attribution, and infringement

This paper examines the copyright challenges raised by generative artificial intelligence (AI). It begins by analyzing the generative process, arguing that the originality of text prompts can serve as a basis for the copyrightability of AI-generated outputs. Accordingly, the study then explores the complex issue of attribution based on the proposed generative process, evaluating the competing claims of users, developers, and data providers, and proposes an ownership model that reflects the degree of creative contribution. Finally, the paper addresses infringement concerns, comparing regulatory approaches in the US, EU and China, and advocates a balanced protection framework that reconciles innovation with copyright safeguards.