EU AI Act: AI labeling rules take effect August 2
What it really says
On August 2, 2026, the transparency obligations under Article 50 of the EU AI Act take effect. From that date, the following rules apply: Chatbots and interactive AI systems must clearly inform users they are interacting with AI, not a human. Deepfakes must be labeled, even without intent to deceive and even if no real person is depicted. AI-generated text on matters of public interest must be marked as such. The European Commission has developed standardized visual labels: 'AI + GENERATED' for fully synthetic content and 'AI + MODIFIED' for manipulated works. AI-generated content must also include machine-readable watermarks or metadata. Systems already on the market before August 2 have until December 2, 2026 to implement machine-readable marking. Penalties for non-compliance can reach up to 15 million euros or 3 percent of worldwide annual turnover. In parallel, the Digital Omnibus was signed on July 8, 2026, deferring high-risk AI deadlines and introducing a new ban on AI-generated non-consensual intimate imagery ('nudifier' apps).
Our assessment
Starting August 2, chatbots must identify themselves as AI, deepfakes must be labeled, and AI-generated text must be marked. Penalties reach up to 15 million euros. The green rating is warranted because this regulation addresses one of the most common AI fears: being deceived by AI without realizing it. Anyone interacting with a chatbot will be informed of that fact. Anyone viewing AI-generated content should be able to recognize it as such. This provides concrete protection against manipulation and deepfake abuse. The Digital Omnibus adds an explicit ban on AI-generated non-consensual intimate imagery. Companies have clear implementation deadlines, and penalties are high enough to enforce compliance.
Relevance for Germany
The labeling requirements apply directly in all EU member states and affect every German company operating chatbots, distributing AI-generated content, or using deepfake technology. Germany's Bundesnetzagentur (Federal Network Agency) serves as the central AI market surveillance authority responsible for enforcement. For German marketing departments, media companies, and technology firms, this means: in eleven days, AI-generated content must be labeled. The regulation is particularly relevant for German election campaigns and political communication, where AI-generated content is increasingly used. The nudifier ban also addresses a growing problem of digital violence in Germany.
Fact check
The transparency obligations under Article 50 of the EU AI Act and the August 2, 2026 enforcement date are confirmed by the European Commission and over ten independent legal analyses. The penalties of up to 15 million euros or 3 percent of global annual turnover come directly from the legislative text. The Digital Omnibus was adopted by the EU Parliament on June 16, approved by the Council on June 29, and signed on July 8, as documented by official EU Parliament and Council press releases. The standardized labels ('AI + GENERATED' and 'AI + MODIFIED') are described in the Commission's guidelines published July 20. The transition period until December 2, 2026 for systems already on the market is consistently reported by Heise, t3n, and multiple law firm analyses.
Source
- • https://www.heise.de/en/news/EU-guidelines-for-AI-labeling-requirements-effective-from-August-11371313.html
- • https://t3n.de/news/ki-kennzeichnungspflicht-marketer-umsetzen-muessen-1752203/
- • https://www.europarl.europa.eu/news/en/press-room/20260427IPR42011/ai-act-deal-on-simplification-measures-ban-on-nudifier-apps
- • https://datamatters.sidley.com/2026/06/24/eu-ai-act-transparency-obligations-preparing-for-compliance-by-2-august-2026/