Coders Say They Already Found Workarounds to Claude’s Invisible Watermarks
Anthropic introduced invisible watermarks in AI-generated content to meet new EU requirements, but within hours of the announcement, people online were already sharing methods to defeat them. The workarounds appeared before the system had any real chance to function in practice.
Why this matters: Watermarks are the EU's main bet for telling AI-generated content from human content. That matters for disinformation, fraud, and basic transparency. If the technique breaks on day one, the policy behind it does not automatically fail, but it does need a harder look. A watermark that coders can strip in an afternoon is not a safety feature. It is a checkbox. The real question is whether regulators are designing rules around what actually works, or around what looks good in legislation.
Who should care: Lawyers · Compliance · General readers · AI governance · Policy
This summary is AI-assisted and may contain errors. It is an original briefing to help you gauge significance quickly — not a reproduction of the source. Always read the linked original before relying on it. See our methodology.