The European Commission has published a code of practice on marking and labelling AI-generated content. It is designed to help organisations prepare for transparency duties under the EU AI Act.

The scope includes deepfakes and public-interest text

The Commission highlights synthetic audio, images and video, as well as certain AI-generated text published to inform the public. Providers of interactive AI systems also face disclosure expectations.

Voluntary code, mandatory duties

Participation in the code is voluntary, but the relevant AI Act transparency obligations apply from 2 August 2026. Following a code does not remove responsibility to meet the law.

Civil-society preparation

Organisations should inventory content tools, preserve provenance metadata where possible, define visible disclosure rules and train staff to distinguish assistive editing from substantially generated public-interest content.