Anthropic will begin watermarking text generated by its AI models, including Claude, to comply with European regulations, the company confirmed in an updated support page. The move follows the EU AI Act’s Transparency Code, which took effect on August 2, 2026, requiring AI companies to mark AI-generated or edited content in a way that other systems can identify.
How the watermark works
According to Anthropic’s support page, all models released after August 2 will automatically include technology to watermark both computer-generated text and files. For files, the company is using the C2PA open standard, a widely adopted technical specification for certifying the origin and history of digital content.
Also read: OpenAI expands Daybreak cyber defense service with new GPT-5.6-Cyber model
The watermark is applied at the model level, meaning it will be present no matter which Claude product or surface the text comes from. This includes the Claude platform API, Claude, Claude Code, Claude Cowork, and Claude Tag.
“Because the watermark is part of the text, it will travel with the text when it’s copied and pasted elsewhere, and may persist through some editing,” the support page reads. The company noted that watermarking will apply across all these products, ensuring consistent traceability.
Also read: Claude-powered agent hacked a gym's booking system to skip the waitlist
It is not yet clear how much editing users need to do to remove the watermark. TechCrunch has asked Anthropic to clarify this and will update the story if the company responds.
Industry-wide shift toward watermarking
Anthropic’s announcement comes as platforms across the AI industry rush to watermark AI-generated content, driven by user backlash and regulatory pressure. Last week, AI music platform Suno said it will mark tracks created on its platform after a spate of legal challenges. Last month, publishing service Substack teamed up with Pangram to flag AI-generated content, with CEO Chris Best calling out “Claudefishing,” a term for people using AI to generate content.
Beyond Anthropic, other major AI companies including Black Forest Labs, Google, Meta, Microsoft, OpenAI, and Synthesia have committed to adhering to the EU’s code. The EU AI Act’s transparency provisions are part of a broader regulatory push to ensure AI-generated content is identifiable, helping to prevent misinformation and protect consumers.
For businesses and developers using Claude’s API, the watermarking is automatic and requires no action. However, the persistence of the watermark through editing raises practical questions for content creators who may want to modify AI-generated text without leaving a trace. Anthropic’s clarification on this point will be important for users who rely on Claude for content production.
The EU AI Act is being phased in over several years, with transparency obligations among the first to apply. As more companies implement watermarking, the industry is moving toward a standard where AI-generated content is clearly labeled, much like nutritional labels on food. This shift could affect how AI-generated content is used in journalism, marketing, and academic work, where provenance is increasingly critical.
This article is for informational purposes only and does not constitute financial or legal advice. The regulatory environment for AI is evolving rapidly, and readers should consult official sources for the latest requirements.
Be the first to comment