A few days ago, Claude's watermark policy sparked widespread attention. Previously, a number of companies, including OpenAI, Anthropic, Google, Meta, Microsoft, etc., had signed the EU's 'Code of Practice on Transparency of AI-Generated Content,' committing to promoting the labeling and detection of AI-generated content. However, Anthropic took a more aggressive approach by not only adding hidden watermarks to all AI-generated text content but also implementing this measure for users worldwide.