Anthropic deploys SynthID-Text watermarking in Claude

Anthropic disclosed on August 14, 2026, that Claude embeds text watermarks using a version of Google DeepMind's SynthID-Text approach, published in Nature in 2024. The watermark modifies word-sampling randomness during generation, leaving a pattern imperceptible to readers but detectable with a key. This is among the first production deployments of AI text watermarking in a major consumer assistant.