Anthropic releases AI content detection tool
Anthropic announced the launch of a new tool designed to detect AI-generated writings by embedding an invisible watermark in texts produced by the "Claude" models. This watermark, which does not affect the meaning or readability of the text, enables the identification of AI content when copied or pasted. It applies to both recent and older "Claude" outputs as part of its efforts to promote transparency, in line with AI regulations in the European Union. The tool is intended to assist publishers and educational institutions in verifying AI usage in writing, while cautioning that extensive modifications might render the watermark undetectable and that it does not definitively prove the text originated from "Claude."
Notice: This Is an AI-Generated Summary
Comments (0)