The protection of AI-generated works under European copyright law : toward adoption of a neighbouring rights approach


University of British Columbia

Date Issued


Document Type



Doctor of Philosophy - PhD




Individuals today create works that outperform their qualitative and quantitative capacities, thanks to generative models such as DALL-E and Midjourney, and sometimes even win awards with these works. These intelligent technologies – incorporated by many into their creative endeavours – throw into question the boundaries of copyright law, built as it is upon human creativity. The unpredictability and autonomous behaviour of AI technologies challenge many of the traditional paradigms of legal systems. The European Union has reacted to these fractures by publishing two Proposals for the AI Act and the AI Liability Directive. The Proposals sketch out a proportionate and future-proof framework for the legal protection of these technologies; but also leave many questions with respect to copyright law unaddressed. This thesis helps to fill this legal gap by analyzing the position of AI-generated works in the EU copyright system, and in so doing, contributes to the European policy-making process. This research comprises five chapters. Chapter One begins with the introduction to the research question. Chapters Two and Three outline the relevant aspects of AI technologies and Union law, respectively. Chapter Four assesses the compatibility of a number of protection models for AI-generated works within the EU copyright system, analyzing whether these technologies can dethrone the human creator, who is at the center of this system, and discard her from the creative process. Under the European copyright system, the author stamps the works with her personal touch, the protection period is based on her life duration, and moral rights are extensions of her personality. This chapter interrogates how the ‘author’, the protagonist of copyright law, can disappear in the creative process, as well as how the concepts of ownership and neighbouring rights could correspondingly evolve. Chapter Five, finally, highlights the need for reform in the legal framework for AI-generated works. It suggests that a sui generis neighbouring right regime would provide much needed transparency and proportionality for AI-generated works, which are, in fact, already a part of copyright systems without explicit acknowledgment of the AI contribution.

Date Available



Attribution-NonCommercial-NoDerivatives 4.0 International




Law, Peter A. Allard School of