Anthropic details how Claude will mark generated content

Anthropic has signed the EU AI Act’s Article 50(2) Code of Practice on Transparency of AI-Generated Content and plans to apply machine-readable marks to content generated by supported Claude models. Claude models launched in the EU on or after August 2, 2026 will support marking at launch, while support for earlier models is still in progress.

The marking system will apply across Claude Platform (API), Claude, Claude Code, Claude Cowork, and Claude Tag, as well as supported deployments through AWS, Google Cloud, or Microsoft Foundry. Text generated by supported models will include imperceptible embedded watermarks, while supported file types such as .svg, .png, and .jpg will include digitally signed provenance metadata based on the Coalition for Content Provenance and Authenticity (C2PA) standard where platform features allow it.

Anthropic says it will support users and third parties in detecting Claude marks, with more technical documentation to come. Detected marks indicate that content may have been processed by Claude, but they do not prove full provenance. Marks may be absent or lost if content came from an older model, was heavily edited, was too short for reliable detection, or had file metadata stripped during conversion, re-saving, screenshots, or other changes.