EU AI Act in Action: Anthropic to Watermark Claude-Generated Content

TuhinChakr · x · 2026-08-13

Anthropic has signed the EU AI Act's Article 50(2) Code of Practice on Transparency of AI-Generated Content. Under these commitments, new Claude models launched in the EU on or after August 2, 2026, will support machine-readable marking from day one.

Generated text will carry embedded watermarks, and generated files will include digitally signed provenance metadata where supported. These marks will apply universally across Claude's API, web, and mobile platforms. Industry observers note that alongside marking AI outputs, future efforts should explore reverse provenance, allowing data owners to verify if their personal content was used for model training.

Related event: Anthropic to Watermark Claude Outputs Under EU AI Act(3 posts)→

Original post →

More from Safety

Safety channel →