The single sharpest fact in one or two punchy sentences. Who did what, where, when, and why it matters. Not a summary of everything — the one thing that makes someone stop scrolling.

Anthropic, the company behind the popular AI model Claude, has announced plans to watermark text generated by its models, including Claude. This move is in response to the enforcement of the European Union's AI Act, specifically the Transparency Code, which took effect on August 2, 2026. The code requires AI companies to clearly mark AI-generated or edited content so other systems can identify them.

The AI Act's Transparency Code is part of the larger EU AI Act, a regulation aimed at ensuring AI systems are used responsibly and safely in the EU. The code specifically targets online content that is generated or edited by AI, requiring companies to provide clear labels or 'watermarks' that identify such content.

Anthropic did not provide details on what the watermarks will look like, but confirmed that the change is in line with the EU AI Act's requirements. This development comes as AI-generated content becomes increasingly prevalent, and AI companies scramble to comply with regulations governing its use.

For a company like Anthropic, which makes a popular AI model used in applications like language translation and content generation, marking its AI-generated content as such is a crucial step in maintaining user trust. The EU AI Act's Transparency Code aims to prevent the spread of misinformation and fake content generated by AI systems.

EU AI Act's Transparency Code took effect on August 2, 2026, and AI companies must now clearly identify AI-generated content. Companies like Anthropic will have to watermark their AI-generated text to comply with the regulations.

Anthropic said it would watermark its AI-generated text to comply with the European regulations. The exact details of how the watermarks will be implemented and how they will look are still unclear. But one thing is certain: the AI Act's Transparency Code is a major push towards greater transparency in AI-generated content.

Key Facts

  • Anthropic confirmed it will watermark AI-generated text to comply with the EU AI Act's Transparency Code.
  • The EU AI Act's Transparency Code requires AI companies to mark AI-generated or edited content.
  • The code takes effect on August 2, 2026.
  • Anthropic's AI model, Claude, will be impacted by the change.
  • The company will mark AI-generated content with a watermark or label that identifies it as such.

This development marks a significant step in the ongoing journey towards greater transparency in AI-generated content. With the EU AI Act's Transparency Code in full effect, companies like Anthropic must adapt quickly to avoid potential penalties and maintain user trust.