Anthropic announced expanded watermarking for its AI text generators. The move arrives as firms scramble to tag synthetic content.

Anthropic will extend support for watermarking to older models, TechCrunch reported.[^1][^2][^4][^6][^7][^8] The hidden pattern will survive downstream processing and can be detected by verification tools.[^2][^3][^5]…

Expanded watermarking across the model stack

Anthropic’s watermark embeds a statistical signal in each token.[^3] The signal is invisible to readers but detectable by a classifier.[^3][^5][^6] The approach mirrors earlier research on steganographic tags.