Digital Sampling and the Erosion of Originality: Rethinking Copyright in the Age of Generative AI

Abstract

As generative AI technologies increasingly replicate artistic styles, voices, and aesthetics by training on vast datasets of existing works, they challenge the foundational assumptions of copyright law. This article explores how AI “sampling” the probabilistic learning and reproduction of creative patterns blurs the legal and ethical boundaries between originality, derivation, and infringement. While AI-generated content is reshaping creative industries through automation and scale, it also risks displacing human creators, undermining labor value, and reinforcing platform monopolies. The article examines legal doctrines of fair use, derivative works, and authorship in light of emerging disputes over AI-generated music, visual art, and literature. It compares regulatory approaches in the United States, European Union, Japan, and Indonesia, revealing a fragmented global landscape with significant implications for developers, artists, and users. Drawing on interdisciplinary sources, it analyses the socio-economic disruptions and proposes a forward-looking framework for legal reform. The study argues that copyright must evolve beyond its human-centric model to accommodate hybrid authorship while ensuring transparency in AI training data, equitable attribution, and protection of human creative agency. In rethinking copyright for the generative era, the article emphasizes a need for inclusive, adaptable, and ethically grounded policy responses that balance innovation with cultural justice.