Meta signs EU transparency code for AI‑generated content ahead of Aug. 2 rules

Meta said Tuesday that it is signing the European Union’s voluntary Code of Practice on Transparency of AI-Generated Content, giving the company a clearer path to showing compliance just days before the bloc’s AI transparency rules begin to apply on Aug. 2.

The move matters because the code is not itself a new legal requirement. Instead, it is a voluntary framework designed to help companies demonstrate compliance with Article 50 of the EU AI Act, formally Regulation (EU) 2024/1689. For companies that sign on, the code offers a recognized route toward a presumption of conformity with those transparency obligations, even though it does not change the underlying law.

Article 50 is the part of the EU AI Act that requires certain AI systems to make clear when content has been generated or manipulated by AI, so users can recognize it more easily. The European Commission’s AI Office published the transparency code on June 10, 2026, and followed it with implementation guidance for Article 50 on July 20. Together, they point companies toward practical approaches such as visible labels or icons and machine-readable provenance markers or metadata, rather than a single mandatory technical standard.

Meta announced the decision in a July 28 post on its newsroom site by Markus Reinisch, Meta’s vice president for public policy in Europe. “Signing the EU AI Act Code of Practice on Transparency of AI‑Generated Content, which we confirmed we will do earlier this week, reflects our commitment to this principle,” the company said. Meta also argued for a more consistent system across platforms and services, saying, “We must avoid a situation where online content is subject to a growing array of different labels and disclosures that end up overwhelming people, while at the same time adding additional regulatory complexity for providers.”

The company tied the announcement to work it has already been doing on labeling AI-made and manipulated media. In April 2024, Meta outlined its approach to identifying and labeling AI-generated content and altered media across Facebook, Instagram and Threads. More recently, on July 7, 2026, Meta previewed a detection tool for images made with Meta AI and said those images could carry a “Content Seal” watermark or provenance signal. “We’re previewing a detection tool that lets you check whether an image carries a Content Seal watermark, providing an initial way to help you better understand if an image was made with Meta AI,” Meta AI said at the time.

Meta’s decision stands out because it had declined to sign the EU’s separate General-Purpose AI Code of Practice in 2025. This week’s announcement concerns a different and narrower code focused specifically on transparency for AI-generated content, not the broader voluntary framework for general-purpose AI. Other large technology companies are also moving to align with the transparency code ahead of the Aug. 2 deadline; Google said on July 24 that it is signing. Meta said it is also working with industry groups including the Partnership on AI and the Coalition for Content Provenance and Authenticity, or C2PA, on interoperable ways to identify AI-generated content.