Anthropic, the company behind the AI model Claude, has announced that it will begin embedding invisible watermarks in all text and sign files generated by Claude. This policy applies globally and will be implemented in new models shipping from August 2026 onward. The watermarks will be detectable using the C2PA standard, and Anthropic plans to provide tools for third-party verification. The move aims to help track and identify AI-generated content. This development matters as it raises questions about the authenticity and ownership of AI-generated materials.