Claude watermarking

Claude watermarking is Anthropic's system for marking AI-generated content from Claude in a machine-readable way. It follows Anthropic signing the EU AI Act Article 50(2) Code of Practice on Transparency of AI-Generated Content. It uses two techniques: an imperceptible watermark woven into generated text, and signed Content Credentials (C2PA) metadata attached to generated files such as PNG or JPEG.

According to Anthropic, the text watermark is part of the text itself. It does not change meaning, quality or readability, travels when text is copied and pasted, and may persist through some editing. It is applied at the model level, so it appears across Claude Platform (API), Claude, Claude Code, Claude Cowork and Claude Tag, and through AWS, Google Cloud and Microsoft Foundry. Claude models launched in the EU on or after August 2, 2026 support marking at launch, and Anthropic is adding it to earlier models. Coverage differs by model and platform. For example, Sonnet 4.6 outputs may not reliably carry the watermark when temperature, top_p or top_k is non-default.

Detection of text watermarks is in private preview. It is available to eligible organizations such as regulators, law enforcement, media, fact-checkers, researchers, educational organizations and EU civil society groups, and to enterprises that must verify marks on their own Claude outputs. Access is requested via a form. C2PA credentials on files can be checked with the free Claude Content Checker. Anthropic says more technical guidance will be published later, and that businesses deploying Claude should assess their own Article 50 obligations.