Anthropic explains how Claude’s invisible text watermarks will work

Anthropic outlined its plan to embed invisible watermarks in text produced by Claude to meet European AI transparency regulations. The watermarking system uses a version of Google DeepMind’s open‑source SynthID‑Text method, which creates detectable patterns based on word‑probability distributions. The feature will be rolled out together with C2PA support for images generated by Claude.

Summary written by Kernelia from the original article by The Verge — AI. The story and its rights belong to its author.