Anthropic Quietly Rolls Out Invisible Watermarks for Claude AI Outputs

AI startup Anthropic has quietly published plans to integrate invisible watermarks into all content generated by its Claude models. This move is designed to bring the company into compliance with the transparency rules of the European Union's AI Act, applying these standards globally across its entire user base. Under the new system, text and code generated by Claude will carry a hidden signature that persists even when copied and pasted elsewhere. Additionally, file outputs will adopt the industry-standard C2PA provenance label. Newer models built after August 2nd already include this feature, and older models are slated to be retrofitted. The rollout has triggered polarized reactions within the tech community. Proponents view it as a necessary step toward digital safety and accountability, whereas critics describe the measure as overreaching and dystopian. This controversy is expected to drive more users toward private, open-source models that offer greater autonomy without embedded tracking.