AIThe Registerabout 3 hours ago
Anthropic unveils AI text watermarking plan
4 min read
Anthropic unveiled a plan to watermark Claude-generated text by altering inconsequential word choices, complying with the EU AI Act. The technique, based on Google DeepMind's SynthID-Text paper, adds a detectable statistical signature. Light editing won't remove the watermark, but a complete rewrite will.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyAnthropic
