EU AI Act Content Labelling Requirements: August 2026 Deadline Guide

The European Union has officially published its AI Content Labelling Playbook, a voluntary Code of Practice designed to assist companies in meeting transparency requirements that will become legally binding across the bloc starting August 2, 2026. The European Commission released the final Code on June 10, outlining practical implementation steps for businesses that develop and deploy generative AI systems to properly mark and label their content outputs.

While the Code itself remains optional, the underlying obligations are mandatory. These requirements are established under Article 50 of the EU AI Act, and from August 2, 2026, they apply to all companies operating in the EU, regardless of whether they sign the Commission's guidance. Signing the Code simply provides businesses with a recognized framework to demonstrate compliance.

📋 What the AI Content Labelling Rules Actually Require

Starting in August 2026, two key elements must be clearly identified:

• Deepfakes and AI-generated or AI-manipulated text published on matters of public interest must carry a visible label
• Users interacting with AI-powered systems, such as customer service chatbots, must be informed they are communicating with a machine

💬 "Europeans have a right to know whether what they see, hear or read has been made or altered by AI, especially when such content can shape public debate," stated Henna Virkkunen, the Commission's Executive Vice-President for Tech Sovereignty, Security and Democracy.

The Commission positions this framework as a mechanism to help users identify AI-created or AI-modified content and reduce opportunities for deception. Virkkunen characterized the Code as a practical implementation pathway that AI providers and deployers can follow before the regulations take effect in August.

🔗 How the Code Divides Responsibilities

The Code distributes obligations across two segments of the AI supply chain:

⚙️ AI Model Providers: Companies that build generative AI models are required to embed machine-readable markers in their output, enabling detection throughout the content distribution chain.

👥 AI Deployers: Companies that implement these models in consumer-facing products are responsible for visible labelling. For public-interest AI-generated text, labelling is mandatory when content is published without human review or editorial oversight.

To ensure practical implementation, the Code relies on open technical standards and a standardized EU icon, designed to provide users with consistent visual identification while preventing businesses from developing fragmented proprietary solutions.

⏰ Implementation Timeline and Next Steps

This Code does not represent the final regulatory framework. The Code is now open for signatures, and the Commission is actively encouraging all providers and deployers to sign. It still requires validation from the Commission and the AI Board, and separate Commission guidelines are forthcoming to clarify legal requirements and address areas the Code does not cover.

Developed by six independent experts with input from more than 180 stakeholders, this represents the first regulatory instrument specifically addressing AI content labelling under the EU AI Act.

⚠️ Critical Deadline: The implementation timeline leaves minimal preparation time. Companies serving European users have less than two months to determine their labelling obligations, implementation methods, and whether to sign the Code. Significant operational details still depend on guidelines the Commission has yet to publish.