Artificial Intelligence and the Transformation of Intellectual Property Protection Rethinking Authorship, Ownership, and Enforcement in the Digital Era

Abstract

The rapid development of artificial intelligence (“AI”) technologies is transforming the manner in which creative, technical, and commercial outputs are generated, distributed, and monetised. Conventional intellectual property (“IP”) frameworks were primarily developed on the assumption that human beings create works, invent technologies, and exercise meaningful control over the resulting intellectual assets. Generative AI, autonomous software systems, machine-learning models, and algorithmically assisted production challenge this assumption by introducing varying degrees of machine involvement between human intention and the final output. This development raises difficult questions concerning authorship, inventorship, ownership, infringement, originality, training data, and enforcement.
This article examines the relationship between AI and intellectual property protection from a doctrinal and policy perspective. It argues that the principal difficulty is not the mere existence of AI-generated outputs but the inadequacy of binary legal categories that distinguish sharply between “human-created” and “machine-created” material. In practice, AI-assisted production exists on a continuum ranging from entirely human-directed activity to substantially autonomous generation. A legally coherent framework must therefore examine the nature and extent of human contribution rather than relying solely on the technological origin of the output.
The article analyses emerging questions concerning copyright in AI-generated works, inventorship in AI-assisted inventions, ownership of training outputs, protection of prompts and datasets, liability arising from infringing content, and contractual allocation of rights. It further proposes a functional framework based on human contribution, control, accountability, and economic investment. Under the proposed framework, copyright and patent systems should continue to preserve meaningful human authorship and inventorship requirements while permitting tailored protection for AI-assisted outputs where identifiable human intellectual contribution is demonstrated. The article concludes that intellectual property law should neither prohibit AI-generated creativity nor grant indiscriminate proprietary rights to machines. Instead, legal doctrine should evolve toward a human-centred but technology-neutral model that allocates rights and responsibilities according to contribution, control, and social policy.