The EU AI Act transparency obligations and Commission enforcement powers apply from 2026-08-02

Sources: European Commission, AI Act, discussion
Summary: The Commission page, last updated 2026-07-31, states that the AI Act's transparency rules come into effect in August 2026 and that from 2026-08-02 the AI Office and Member State authorities are responsible for implementing, supervising, and enforcing the regulation. The transparency obligations require providers of generative AI to ensure AI-generated content is identifiable, and require deep fakes and text published to inform the public on matters of public interest to be clearly and visibly labelled. The AI Office holds enforcement powers over general-purpose AI models, including requesting technical documentation, evaluating models, requiring corrective measures, and issuing fines, and guidelines on transparency obligations for providers and deployers were published 2026-07-20.
Why it matters: A team shipping a generative feature into the EU market has to make its output machine-identifiable and label deep fakes and public-interest text from 2026-08-02, and the body that can demand technical documentation and evaluate models holds that power from the same date.
Follow-up: The AI Omnibus amending regulation entered into force 2026-07-27 and moved the high-risk rules to 2027-12-02 for Annex III use cases and to 2028-08-02 for Annex I regulated products, so track those two dates separately.