Meta Signs EU AI Transparency Code Days Before Deadline

Meta has signed the European Union's Code of Practice on Transparency of AI-Generated Content, placing Facebook, Instagram, and Threads inside a voluntary framework just five days before Article 50 of the AI Act becomes legally binding on August 2, 2026.

What the Code Requires

The transparency code, published by the European Commission on June 10, establishes concrete technical obligations for AI content labeling. Under Section 1, providers like Meta must implement machine-readable marking with at least two layers — digitally signed metadata plus imperceptible watermarking — on AI-generated audio, images, and video. Free-form text longer than 200 tokens must also be watermarked. Signatories are required to offer a free detection tool to users, researchers, and civil society groups, with content deleted immediately after processing.

Section 2 covers deployers — meaning brands, publishers, and anyone publishing AI-generated content publicly. Deepfakes and AI-generated text on matters of public interest must carry visible labels using the EU's standard icon set or an equivalent design, placed where users see it before or at first exposure.

A Calculated Move

Meta's decision is strategically significant. The company famously declined to sign the EU's General-Purpose AI Code of Practice in 2025, citing legal uncertainties. This new transparency code is narrower — focused solely on content labeling, not model governance — and signing it gives Meta a recognized route to demonstrate compliance rather than defending ad-hoc measures before market surveillance authorities across 27 member states.

The stakes are real: non-compliance with Article 50 carries fines of up to €15 million or 3 percent of worldwide annual turnover, whichever is higher.

The Bigger Picture

Meta has been building toward this since February 2024, when it announced its approach to labeling AI-generated content. In July 2026, the company previewed a detection tool that checks whether images carry a "Content Seal" provenance watermark. The company is also active in the Coalition for Content Provenance and Authenticity (C2PA) and the Partnership on AI, pushing for interoperable standards.

"Getting this right is critically important," Meta's VP of Public Policy Europe, Markus Reinisch, wrote in the announcement. "It means transparency measures that provide clarity, rather than more confusion. We must avoid a situation where online content is subject to a growing array of different labels and disclosures that end up overwhelming people."

Google confirmed on July 24 that it is also signing the transparency code. The broader industry alignment suggests that — at least on content labeling — major platforms are converging on a common playbook, even as debates over open-weight models and frontier AI governance continue to fracture the industry.