Skip to content
Feature deep-dive

The EU’s AI Labeling Rule Is Already in Force. Here’s What Creative Teams Actually Have to Disclose

Article 50 of the EU AI Act took effect on August 2, 2026, requiring visible disclosure of realistic deepfakes and machine-readable marking of AI-generated media. It reaches companies outside the EU whose content is seen by people in the EU, but the Commission itself says not all AI content needs a label, and creative work gets a lighter-touch rule.

2 min read

The EU’s AI Labeling Rule Is Already in Force. Here’s What Creative Teams Actually Have to Disclose

Article 50 of the EU AI Act took effect on August 2, 2026, turning AI content labeling from a best practice into a legal requirement. The rules split the work into two jobs. Providers of AI tools must mark generated audio, image, video, and text in machine-readable form, while the people who publish the content must clearly disclose deepfakes, meaning generated or manipulated media that resembles real people, places, or events and could falsely appear authentic. The act applies to companies outside the EU that create AI-generated content visible to people within the EU, so a studio or brand elsewhere is not automatically exempt. Two points work in a creative team’s favor. The European Commission’s own guidance states that not all AI-generated or manipulated content needs to be labelled, and for evidently artistic, creative, satirical, or fictional work, disclosure only has to inform the audience without disrupting the presentation or enjoyment of the work. The Commission has also released optional EU icons for labeling, and tools already on the market have until December 2, 2026 to meet the machine-marking requirement.

In practice, this turns on a simple question for each asset: could a viewer mistake it for something real? A stylized animated spot or an obviously fictional film sits very differently from a realistic synthetic person presented as a real customer. Teams that publish AI-assisted work into Europe should build a few habits now. Keep an inventory of the AI tools used, record where each asset came from and how it was edited, and add a review step before publishing realistic synthetic media. These records also help with client questions about how a piece was made. Reported penalties reach up to EUR 15 million, which is reason enough to treat this as a workflow issue rather than an afterthought. This is general information and not legal advice, so projects with real exposure should be checked with a qualified adviser. The broader point holds either way: transparent, well-documented AI use is becoming part of professional credibility.

Keep reading

More from the blog

View all