EU Commission Releases AI Labelling Code Before Aug 2
Key insights
- The Code of Practice is voluntary, but the AI Act transparency obligations it supports become enforceable on August 2, 2026.
- Three categories require labelling: deepfakes, AI-generated or manipulated text on matters of public interest, and interactive AI systems such as chatbots.
- The European Commission released EU icons for labelling AI-generated content alongside the code to reduce implementation friction for providers.
Why this matters
Generative AI providers and deployers operating in the EU face an August 2, 2026 transparency compliance deadline, and this code gives them a concrete, Commission-endorsed framework to demonstrate they have met it. The three-category labelling structure covering deepfakes, public-interest AI text, and chatbot interactions sets a specification that other jurisdictions will likely reference as they build their own AI disclosure requirements. The release of official EU icons and supplementary FAQs substantially reduces design and legal ambiguity for product teams shipping AI features into European markets.
Summary
The European Commission published a voluntary Code of Practice on June 10, giving generative AI providers and deployers a structured path to meet AI Act transparency obligations before August 2, 2026.
The code covers three categories: deepfakes, AI-generated or manipulated text on matters of public interest, and interactive AI systems like chatbots. The Commission also released EU icons specifically for labelling AI-generated content alongside implementation guidance and FAQs.
Essentially: the Commission provides a voluntary compliance pathway for providers and deployers ahead of the August 2 deadline.
- Deepfakes, AI-altered public-interest text, and chatbot interactions all require clear identification.
- EU icons for labelling AI-generated content are available to ease implementation.
- The code is voluntary; the AI Act transparency obligations it supports are not.
The stated goal is helping users recognise when content has been generated or altered by AI, reducing the risk of deception and manipulation.
Potential risks and opportunities
Risks
- Providers who do not sign the code and lack an alternative transparency compliance mechanism face enforcement exposure under the AI Act starting August 2, 2026.
- The undefined boundary of 'matters of public interest' for AI-generated text creates legal uncertainty for news publishers and platform operators deploying generative AI tools in the EU.
- Inconsistent or low-quality adoption of EU labelling icons across providers could undermine the user recognition the regulation is designed to establish, weakening its stated deception-reduction goal.
Opportunities
- Compliance tooling vendors building AI content provenance and watermarking solutions gain a clear, Commission-backed specification to target for EU market readiness before August 2.
- Publishers and platform operators who adopt EU icons and labelling practices early can differentiate on transparency and build user trust ahead of slower-moving competitors.
- The Directorate-General for Communications Networks, Content and Technology's framework signals appetite for further AI disclosure standards, opening space for legal consultancies and standardization bodies to develop sector-specific guidance.
What we don't know yet
- Whether signing the voluntary code creates a legal safe harbor against AI Act enforcement actions, or merely signals good-faith effort, is not addressed in the published materials.
- How 'matters of public interest' will be defined and enforced for AI-generated text labelling remains unresolved before the August 2 compliance deadline.
- The code does not clarify whether providers who meet transparency obligations through alternative means satisfy the AI Act requirements as fully as those who sign the code.
Shared on Bluesky by 1 AI expert
Originally reported by digital-strategy.ec.europa.eu
Read the original article →Original headline: European Commission Publishes Final Code of Practice on AI Content Labelling — Deepfakes and AI-Generated Text Must Be Marked by August 2