Anthropic is watermarking text from new Claude models under EU transparency code

@AnthropicAI has begun embedding invisible watermarks in text produced by its @claudeai models, rolling out the feature in response to new European Union transparency rules and applying it worldwide.

What the Watermark Does and Does Not Prove

Claude models launched on or after August 2, 2026 support marking at launch. When a supported Claude model generates text, it weaves an imperceptible watermark directly into the text itself. It does not change the meaning, quality, or readability of the response, and because the watermark is part of the text, it will travel with the text when copied and pasted elsewhere, potentially persisting through some editing.

The distinction between processing and authorship matters. A detected mark shows that Claude handled the content, not necessarily that Claude wrote it from scratch. Users frequently paste their own work into the model for editing or rewriting, meaning a watermark is not confirmation of full AI authorship. Anthropic says it will publish technical details for detecting its watermarks, and those tools will help users and third parties check content for supported Claude marks, though detection tooling has not yet been released publicly.

For files, the company is using the C2PA open standard. Anthropic noted that a file's marking can be stripped through format conversion, re-saving, screenshots, or similar processes, meaning a document or image that once carried a Claude mark may no longer show any trace of it after being edited or converted. Anthropic is also working to add marking support to Claude models released before August 2, 2026, and says it will update its Help Center as that becomes available.

Global Rollout Driven by EU Regulation

The timing traces directly to the EU AI Act, which requires that AI systems creating synthetic content mark their outputs as artificially generated under Article 50. The transparency obligations under Article 50 took effect on August 2, 2026, with non-compliance potentially triggering fines of up to 15 million euros or 3 percent of total global annual turnover, whichever is higher.

Anthropic also signed the EU's Article 50(2) Code of Practice on Transparency of AI-Generated Content, which the European Commission and AI Board have assessed as an adequate compliance demonstration, shifting the evidentiary burden toward regulators. The EU regulation applies to AI systems used within the bloc, but Anthropic applied the mark globally anyway, with its Help Center stating that marking will apply to output from supported models wherever Claude is offered, worldwide. This is another example of how the EU's standard of compliance can have ripple effects on global AI norms.

Claude markings cover output from supported models everywhere Claude is used, including Claude Platform (API), Claude, Claude Code, Claude Cowork, and Claude Tag.

Sources:
Anthropic Help Center: How Claude marks AI-generated content
TechCrunch: Anthropic says it will watermark text generated by its AI models
Euronews: Anthropic to watermark Claude's output worldwide