EU Code of Practice on Transparency of AI-Generated Content Takes Effect August 2
Key Takeaways
- ▸The Code of Practice for AI transparency compliance is now legally applicable as of August 2, 2026, with approximately 190 organizations already signed up to provide a unified EU framework.
- ▸Signatories commit to marking/detecting AI-generated content (providers) and labeling deepfakes and AI-generated text (deployers), using standardized EU-provided icons for consistency.
- ▸Signatories gain legal predictability and reduced administrative burden across all EU member states, while non-signatories face individual compliance verification by market surveillance authorities.
Summary
The European Union has activated the Code of Practice on Transparency of AI-Generated Content, marking a significant milestone in implementing the AI Act's transparency requirements starting August 2, 2026. This voluntary yet EU-endorsed framework establishes unified standards for marking, labeling, and detecting AI-generated materials, aiming to combat deception and manipulation in the information ecosystem. Approximately 190 organizations—including AI providers, deployers, civil society groups, and academic experts—have already committed to the code, signaling strong adoption across the bloc.
The code operates through two complementary mechanisms: providers must implement marking and detection systems for AI-generated or manipulated content, while deployers must label deepfakes and AI-generated text using standardized EU icons. Signatories benefit from legal certainty and simplified compliance, avoiding fragmented assessment by individual member states' market surveillance authorities. The European Commission has officially endorsed the code as an adequate demonstration of compliance with Article 50 of the AI Act. Non-signatories must independently demonstrate the adequacy of their compliance measures to respective national authorities, creating regulatory burden and unpredictability.
- The framework emerged from a multi-stakeholder process including AI providers, civil society, academics, and international observers, ensuring practical and proportionate implementation of transparency obligations.
