EU AI Act Sets AI‑Generated Content Labeling Rules Effective August 2026

The European Union’s AI Act will make transparency obligations binding on 2 August 2026. organisations must disclose when users interact with AI systems such as chatbots, and certain AI‑generated content – especially deepfakes and material of public interest – must carry visible labels indicating its artificial origin.

The European Commission released a voluntary Code of Practice on 10 June, a “playbook” that advises developers to embed machine‑readable markers (watermarks, metadata) in generated outputs and urges deployers to ensure those markers survive through publishing pipelines and are presented to end‑users. While the code is optional, the legal requirements are not, and non‑compliance can attract fines of up to €35 million or 7 % of worldwide annual turnover.

Implementation timelines are split: the core transparency duties begin in August 2026, while the specific watermarking requirement is postponed to December 2026. Regulators stress that companies need robust inventory, documentation and risk‑classification of their AI systems to meet the new rules, and that compliance will become a standard part of due‑diligence for investors and partners.