Meta is Signing the EU AI Act Code of Practice on Transparency of AI-Generated Content

Meta is signing the EU AI Act Code of Practice on Transparency of AI-Generated Content, according to the Meta Newsroom. The move formalizes work Meta has already framed as necessary as AI-generated media grows more photorealistic: people need tools that help them identify when content was produced by AI rather than captured or created by a human.

The company’s public line on this goes back to February 2024, when Meta announced an approach to identifying and labeling AI-generated content. Signing the Code of Practice puts that labeling and identification work under a shared EU-oriented transparency standard rather than treating it only as an internal product policy. The focus is on disclosure and detectability of synthetic media, not on banning generation tools themselves.

For engineers and product builders, this matters because transparency of AI-generated content is becoming a compliance and product surface, not only a trust-and-safety talking point. Pipelines that produce images, video, or other synthetic media for consumer platforms will increasingly be expected to carry labels, provenance signals, or other identifiers that end users and platforms can read. Teams shipping generative features into EU-facing products should treat labeling and content identification as first-class requirements in the generation and distribution path.

In market terms, Meta is aligning with an EU regulatory instrument while other large platforms face the same pressure to show how they mark AI media. Photorealistic generation raises the cost of leaving synthetic content unmarked: users cannot reliably tell machine-made media by eye alone. By signing onto the Code of Practice, Meta is signaling that its identification and labeling approach will be measured against a common transparency baseline rather than only company-specific rules.

What to watch next is how Meta implements identification and labeling under the Code in product flows, and how those signals show up for creators and consumers when content is uploaded, shared, or recommended. Builders should track whether similar transparency commitments force consistent labeling across generation APIs, upload paths, and feed surfaces—and whether Meta’s post–February 2024 approach is extended or hardened as photorealistic AI media keeps spreading.