EU AI Act Sets Strict Transparency Rules for AI-Generated Content

New requirements under the EU AI Act will introduce binding transparency obligations for AI-generated and manipulated content from August 2026. These measures are designed to ensure users are informed when interacting with AI and to support detection of synthetic media.

Article 50 of the EU AI Act mandates that providers of AI systems must implement technical features that notify users when they are engaging with AI. In addition, providers are required to include machine-readable markers in content produced or altered by AI to facilitate identification by automated tools.

Deployers of AI systems must also apply clear labels to deepfakes and AI-generated or manipulated text, especially when such content addresses matters of public interest and has not been subject to human review or editorial oversight. These requirements aim to improve transparency and accountability in the use of AI for generating or modifying digital content.

Manual disclaimers, such as simple captions or generic notices, are unlikely to fulfill the technical and machine-readable marking standards set by Article 50. Analysis from industry sources indicates that effective compliance will require more robust controls than basic policy statements or design-time assurances.

What the numbers show

• Article 50 becomes enforceable on 2 August 2026.
• Non-compliance can result in fines up to €15 million or 3% of global turnover.
• The Code of Practice on Transparency of AI-Generated Content is a voluntary compliance tool.

National market surveillance authorities, the EU AI Office, and the European Data Protection Supervisor (for EU institutions) are responsible for enforcing these transparency rules. Organizations are increasingly approaching AI governance as an operational matter, reflecting the binding nature of the new regulatory framework.

Expert analysis suggests that AI compliance programs often fall short when they rely solely on written policies or manual disclaimers. Instead, ongoing runtime governance and enforceable technical controls are necessary to meet the EU AI Act’s requirements.

The European Commission and the AI Board have assessed the Code of Practice on Transparency of AI-Generated Content as an adequate voluntary resource to help providers and deployers demonstrate compliance with Article 50. This code offers guidance but does not replace the mandatory technical obligations set by the Act.

Industry reaction

Anthropic announced that future versions of its Claude AI models will include watermarks to align with the EU AI Act’s marking requirements. The company stated that this step is intended to support compliance with the new transparency obligations.

Industry reports indicate that organizations are adapting their AI governance strategies in response to the operational and legal demands introduced by the EU AI Act. These changes reflect a broader shift toward enforceable technical measures over policy-based approaches.

* This article is based on publicly available information at the time of writing.