Generative Artificial Intelligence-Related Copyright Regulation Concerns, Issues, and Policies

Abstract

This paper provides a systematic literature review of studies investigating generative artificial intelligence (AI)-related copyright regulation concerns, issues, and policies. Our analysis highlights copyright doctrine application to generative AI in terms of data security, infringement, and fair use. The inspected databases were the Web of Science, Scopus, and ProQuest. For original and review article screening and quality assessment, we leveraged the following review software systems: Abstrackr, CADIMA, DistillerSR, EPPI Reviewer, MMAT, ROBIS, and SRDR+. Dimensions and VOSviewer were harnessed for bibliometric mapping and layout algorithms with regard to data visualization and analysis. PRISMA was the reporting quality assessment tool. Our results show that copyrighted data can be deployed in generative AI system training fairly and coherently, thus decreasing copyright-infringing content.