How Claude's text watermarking works
Anthropic
Anthropic is implementing text watermarking in future Claude models to comply with the EU AI Act without affecting output quality.Summary
Anthropic is introducing text watermarking for future Claude models to comply with the EU AI Act, using a SynthID-Text-based approach that leaves an undetectable statistical pattern in generated words. This watermarking method has no practical impact on quality, creativity, or readability, does not add extra tokens or cost, and preserves user privacy by containing no identifying information. While effective for longer passages, the watermark's presence is minimal in factual texts, code, or lightly edited human writing.
(Source：Anthropic)