A statutory compensation model for AI training based on the copy levies payment framework

Abstract

The article addresses the lack of adequate legal mechanisms regulating the use of copyrighted works in the training of artificial intelligence (AI) algorithms, particularly in the context of copyright law. The author argues that current licensing solutions, especially those based on collective rights management organizations are insufficient given the scale and nature of AI training operations. In response, the article proposes the adoption of a statutory compensation model inspired by the copy levies system used in the European Union. Under this framework, AI developers would be required to pay a fee for using protected content, with the proceeds distributed by authorized organizations to the rightsholders. The article analyzes both the advantages and limitations of this model, highlighting its potential as a tool for balancing the interests of creators and the innovation sector. The author contends that only a fair and systemic compensation mechanism can ensure the sustainable development of both AI technologies and creative industries.

Keywords

• artificial intelligence (ai)
• ai model training
• copy levies
• copyright law
• statutory compensation