Anthropic Is Quietly Watermarking Every Claude AI Output. Builders Are Already Trying to Break It
60-second summary
Anthropic is quietly watermarking every output from its newest Claude AI models with an invisible, machine-readable code, sparking concerns about data ownership and control. The exact method remains undisclosed, and builders are already attempting to break the watermark, raising questions about the implications for AI-generated content and its potential to be traced back to its creators.
Anthropic is weaving an invisible, machine-readable watermark into every word its newest Claude models write—and it hasn't said how.