Anthropic to Embed Invisible Watermarks in Claude Outputs From August 2026
Anthropic has signed the EU AI Act's Article 50(2) Code of Practice on Transparency, committing to machine-readable marking of AI-generated content. Starting August 2, 2026, all new Claude models will embed an imperceptible watermark in generated text and cryptographically signed C2PA metadata in supported image files. The marking applies globally — not just in the EU — across the Claude Platform API, Claude Code, Claude Cowork, and Claude Tag, including deployments via AWS, Google Cloud, and Microsoft Foundry. Watermarks cannot be disabled and are embedded at the model level, meaning they can survive copy-paste across platforms. Importantly, the presence of a mark does not confirm Claude authored the content, nor does its absence prove human authorship.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.
Discussion (0)
Log in to join the discussion and vote.
Log in