Fair Remuneration in Times of Generative AI - A Reality Check for the Dutch Music Sector
Among artists and the creative industries, generative AI (GenAI) has triggered a heated scientific and public debate on issues ranging from parasitic use and misappropriation of copyrighted works to accountability, liability and remuneration. This project addresses a critical aspect of this discourse within the Dutch music ecosystem: how to ensure that artists and producers in the music industry receive remuneration for the use of musical works in AI training?
A well-functioning licensing and remuneration system not only satisfies compensation interests of artists and rightsholders, but also gives AI developers access to diverse training repertoires and allows them to develop generative models that are unbiased.
The European Artificial Intelligence Act (AIA) and the 2019 Directive on Copyright in the Digital Single Market (CDSMD) provide the legal framework for Dutch remuneration claims. The CDSMD grants rightsholders the option to prohibit the use of their works for AI training purposes, while the AIA obliges AI developers to provide summaries of works used for AI training. Together, these regulations empower rightsholders to control their works’ use, but it is unclear how fair remuneration and appropriate AI revenue distribution could be achieved under these frameworks.
Against this background, the researchers in this project will conduct a reality check to clarify how these EU regulations could be implemented to ensure fair remuneration for artists and rightsholders. This project brings together Explainable AI, sociological research, and legal expertise.
Three non-academic partners will support the project:
The International Federation of the Phonographic Industry (IFPI), representing the global music industry worldwide and offering expertise on individual rightsholder perspectives;
BumaStemra, the collective rights organisation for Dutch composers and music publishers;
The Kunstenbond, an artists’ union offering access to individual musicians and text writers.
Project team:
University of Amsterdam: Dr. John Ashley Burgoyne (Musicology, Faculty of Humanities), Prof. Dr. Stephanie Steinmetz (Sociology, Faculty of Social and Behavioural Sciences), Prof. Dr. Martin Senftleben (Copyright, Faculty of Law)
The project has three central objectives, including policy proposals. See especially the third point:
● quantify the influence of specific training examples on GenAI music compositions, analyzing their encoding paths within the 'brain' of GenAI systems;
● investigate how GenAI affects working conditions for artists and rightsholders; and
● develop policy recommendations and guidelines for fair remuneration and licensing, balancing the needs of artists, rightsholders, and AI developers by combining insights from computational, musicological, sociological, and legal perspectives
The results will be published on open access and made available on the project website and shared widely with music makers, producers, distributors etc. in the chain, policy makers, and academia.
More to read
