Creators: Transparency and Consensus Key to Ethical Use of Artificial Intelligence

Authors and performers support the development of digital technology, with which they can enrich their works of art and reach their audience even more easily. This also includes the development of artificial intelligence, but in this area it is necessary to ensure that the creators of this technology, who are increasingly using the works and creations of authors and performers to complete the content and develop their applications, obtain the consent of the creators for this. By not asking for consent, they violate the copyright of authors and performers at all levels, six umbrella associations of authors and performers wrote in a letter to decision-makers in the EU.

As pointed out by the European Directors Association (FERA), the Federation of Screenwriters of Europe (FSE), the International Federation of Actors (FIA), the European Composers and Lyricists Association (ECSA), the European Writers’ Council (EWC) and the International Federation of Musicians (FIM), creators and umbrella organizations do not oppose the development of artificial intelligence, but even support the access of creators of these systems to various works and contents of creators. However, in order to use the works (texts, images, sound, videos, video games) it is absolutely necessary to obtain the consent of the rights holders first.

The European Commission is preparing new legislation in the field of artificial intelligence, which contains some safeguards that should ensure respect for copyright. This mainly concerns Article 52 of the proposed act on artificial intelligence, which requires operators of such systems to disclose that the content was artificially created or changed. However, the general approach of the Council of the EU now includes an exception to this principle, namely everywhere “where the content is part of a creative, satirical, artistic or fictional work or program” (paragraph three of Article 52 of the Act on Artificial Intelligence, recital 70).

Authors and performers agree with this proposal and call on all EU policy makers involved to oppose such ambiguous and harmful text that can be used against the rights and interests of authors, performers and citizens. “We oppose any need to exclude this content from transparency requirements and any secrecy related to the use of artificial intelligence. On the contrary, we strongly support maximum transparency in all aspects of the development, production and distribution of creative works using artificial intelligence technologies. Providing (unjustified) exceptions to lack of transparency would have serious consequences for our economic and moral rights, as well as for the ability to exercise these rights,” the umbrella organizations of creators emphasize on behalf of the creators.

Therefore, they also call for clarification of the approach announced by the Legal Affairs Committee of the European Parliament (JURI) in its amendments to Article 52(3). The petition’s signatories demand that all possible “deep fakes” be covered in the proposed legislation. In addition, it is imperative that managers of artificial intelligence systems, when it creates or alters content that is part of a creative, artistic or fictional audiovisual work, with the consent of all those concerned. Therefore, representatives of authors and performers call on all decision-makers in the EU to put the concept of transparency and consent of authors and performers for the use of their works at the center of all their initiatives related to the use of artificial intelligence.