The European Commission's AI Office, working with national authorities, began enforcing the EU AI Act's transparency obligations on August 2, the date the rules' Article 50 requirements took effect across the bloc. The Commission also opened its first formal enforcement powers under the Act on the same day, even as separate high-risk system obligations that were originally due on the same date were pushed back to 2027 and 2028.

What the transparency rules require

Chatbots and other interactive AI systems must now tell users they are communicating with AI rather than a human. Deepfakes — images, video or audio edited or generated by AI — must be labeled, and AI-generated or altered content more broadly must carry machine-readable marks so it can be detected. Alongside the enforcement start, the Commission published a first list of more than 180 organizations that have signed its voluntary Code of Practice on transparency of AI-generated content, and issued guidelines clarifying which providers and deployers the obligations cover.

Enforcement powers, and what got delayed

The AI Office can now request information and documentation from providers, obtain access to models for evaluation, require corrective or risk-mitigation measures, and impose fines of up to the greater of €15 million or 3% of a provider's global annual turnover. That enforcement authority arrives alongside the Digital Omnibus on AI, a package the Commission agreed on July 8 that postponed the Act's high-risk system obligations — originally due August 2, 2026 — to December 2, 2027 for stand-alone high-risk systems and August 2, 2028 for AI embedded in regulated products. Implementation of those provisions had stalled in part because member states had been slow to designate national competent authorities and the harmonized technical standards the compliance regime depends on were not yet finished.

Why it matters

The split outcome — transparency rules live and enforceable now, high-risk obligations deferred by one to two years — gives AI developers and deployers operating in the EU a clearer near-term compliance target than the uncertainty of the past several months, while acknowledging that the bloc's regulatory infrastructure for the harder high-risk category isn't ready. For companies shipping chatbots, image generators or other AI-facing products in Europe, the disclosure and labeling requirements are now a live legal obligation with real fines attached, not a future deadline to plan around.