Why Anthropic’s AI watermark is going further than its rivals — for now
Anthropic has implemented a robust AI watermark for its Claude model that persists through edits and translations, sparking debate over its necessity and impact. Critics argue the implementation is overly aggressive compared to EU regulatory requirements, potentially hindering legitimate user workflows.
Why it matters
This highlights the tension between AI transparency, regulatory compliance, and user experience in the rapidly evolving generative AI landscape.
Anthropic CEO Dario Amodei and OpenAI CEO Sam Altman. Chris Ratcliffe/Bloomberg via Getty Images; Eric Lee/Bloomberg via Getty Images Anthropic's push to label AI-generated text may be going further than what its rivals are doing. Its watermark can survive some edits and has prompted backlash. "Anthropic is using a sledgehammer that hits simple proofreading and translation," one policy expert said. Anthropic's new watermark for AI-generated text struck a nerve for some Claude users — but did Europe's laws require it to go as far as it did? A Business Insider review of the law, supported by interviews with AI and policy experts, suggests Anthropic's approach may go beyond what the EU requires. Under the European Union's Article 50(2), AI companies do not have to mark AI outputs when the technology is used for standard editing or does not substantially change a user's original words or meaning.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in