AI Models Are Watermarking Text—Will You Notice?

Anthropic announced that future Claude models will include cryptographic watermarks in their generated text to identify content as AI-produced. This move aligns the company with Google, which already applies similar watermarking to its Gemini models. These technical signals aim to improve transparency by allowing platforms to distinguish between human-written and machine-generated content. As major developers adopt these standards, it marks a shift toward standardized methods for tracking the origin of digital text.