VITraDes

Verbundprojekt: Vertrauenswürdige Information und Transparenz zu Algorithmen und lernenden Systemen im Journalismus - VITraDes

01/04/2026 to 31/03/2029

In order to counter disinformation through journalistic professionalism and quality, journalism is subject to
special transparency and labeling requirements, particularly when using artificial intelligence (AI). Labeling
and transparency requirements also play an important role in combating disinformation in the European
Union's AI Regulation. While the AI Regulation generally requires the labeling of AI-generated content,
exceptions are provided for art and journalism. These freedoms, or uncertainties, in dealing with
transparency requirements raise the question of how labeling and transparency standards for AI can be
implemented in journalism in a way that is meaningful, informative, and appropriate for the target audience.
In the interests of journalistic independence, but also to ensure that labeling is standardized and
transparent, these requirements should be developed independently as part of media self-regulation.
Especially for professional journalism and to protect public communication spaces from disinformation, it is
essential to safeguard trust in media reporting through transparent voluntary commitments.VITraDes is
developing the theoretical foundations for AI labeling and transparency in journalism in an interdisciplinary
manner (media ethics, journalism, law) and is creating practical guidelines in a participatory process. From
a media ethics perspective, the focus is on AI ethics guidelines, which are being developed as part of the
ethical commitments being made by media companies and editorial offices. The ethical foundations for
transparency requirements and their specific practical implications are being developed. The subproject also is responsible for coordinating the overall project.