The dilemma of authorship for AI-generated work in the  EU and US. A comparative study of the notions of ‘human input’ and ‘author’s own intellectual creation’ in the assessment of authorship for copyright protection of AI-generated work

Abstract: With the technological advancements of the 21st century, there has been a significant increase of AI-generated work in both the EU and the US. However, authorship of AI-generated works has been a contested issue as it challenges the traditional view of creations being associated with human beings. Currently, there is no specific copyright legislation that regulates the copyrightability of AI-generated works. Consequently, courts in both jurisdictions have to rely on existing copyright legislation in their assessments of authorship. The thesis analyzes the existing copyright legislation in both jurisdictions and, most compellingly, argues that authorship is considered a human trait in both the EU and the US. However, it will also pinpoint that there is a lack of case law in the EU that specifically addresses AI-generated work, as opposed to the US. The thesis also concludes that the existing copyright legislation in both jurisdictions is not suitable for the assessment of authorship of AI-generated work. On that note, the thesis also concludes that, while the legislation is not suited, it can still be argued to be appropriate for the courts to use.