Anthropic to watermark AI-generated content
Anthropic will begin labeling AI-generated content created by Claude, Claude Code, and the company’s API, as well as Claude models via third-party services, according to The Register . Text generated by future Claude models will be given an invisible watermark. According to Anthropic, the watermark is embedded directly into the generated text without affecting its meaning or readability. For…
Anthropic, the AI company behind the popular conversational AI Claude, has announced plans to introduce invisible watermarks to identify AI-generated content, according to a report by The Register. This move is aimed at ensuring transparency and compliance with the EU's stringent AI Regulation. The watermarks will be embedded directly into the generated text, leaving the content's meaning and readability intact.
For non-text files, Anthropic will utilize signed metadata in line with the C2PA standard. This initiative is part of Anthropic's broader strategy to adhere to transparency requirements concerning AI-generated material. However, the company has acknowledged the limitations of the new labeling system. While a label indicates the text was generated by Claude, its absence doesn't conclusively prove the material wasn't AI-generated.
Importantly, the global implementation of this feature applies to all Claude users, regardless of their location or the service through which they access the AI.
Written by urgent.news from Computerworld's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
- Claude now embeds invisible watermarks in AI-generated text nairametrics.com
- Claude to start watermarking AI-generated content mashable.com
