AIHacker Noonabout 3 hours ago
Anthropic Watermarks Claude Output, Flagging Human Edits
12 min read
Anthropic began watermarking Claude's output this week, marking text invisibly when copied. The detection flag catches writers like Peter Harrell and Maya Sen whose human drafts were edited by Claude, while paraphrased AI filler passes clean. Anthropic's documentation admits the mark is not proof of AI authorship.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyAnthropic
Summary by ByteBrief
