Anthropic Adds Invisible Watermarks to Identify AI-Generated Text Globally
Anthropic has introduced an invisible statistical watermark to text generated by its AI system, Claude. This hidden pattern, undetectable to readers, allows identification of AI-generated content even after minor edits. The watermarking aligns with transparency requirements under the European Union's AI Act and is being implemented globally. Additionally, the system supports provenance data for images, aiding in detecting tampering. Anthropic states the watermark does not affect text quality or readability and is designed to indicate AI involvement without revealing user identity.
First-hand measurement across 2 sources
We measured how 2 outlets covered this story. No outlet gave this story a measurable political slant — there is no left–right reading to report. Overall sentiment is neutral (51/100). Lens Score 40/100.
Outlets measured: wion, ndtv. See how each one headlined and framed the same story in the source comparison below.
AI Analysis
Sentiment was consistent across outlets (50–52/100), indicating broadly factual reporting rather than editorialising.
Coverage timeline
ndtv broke this story on 14 Aug, 10:51 am. Other outlets followed.
- 1