Anthropic Applies Statistical Signal-Based Watermark to Claude

Anthropic is applying a machine-readable statistical signal-based watermark to content generated by Claude. This method determines Claude's involvement by using signals left during the generation process instead of hidden characters or codes. Anthropic has signed the EU AI Act's Article 50 implementation regulation on 'Transparency of AI-Generated Content,' stating that Claude models released after August 2, 2026, will support this marking feature. The applications include the Claude platform (API), Claude, Claude Code, Claude Co-work, and Claude Tags. Anthropic explained that this marking does not change the meaning, quality, or readability of the responses. Files will include signed source metadata, adhering to the C2PA standard. Google's DeepMind's SynthID text also addresses similar issues, generating watermarks that are invisible to humans by adjusting probability values. Anthropic noted that detection results only indicate a possibility that the content was processed by Claude, without confirming the entire source. According to Blockbits, some Claude users canceled their subscriptions after the watermark was introduced, but no overall increase in cancellations has been observed. This measure relates to the competition among platforms surrounding AI content certification.

This content is provided for general informational purposes only and doesn't constitute financial, investment, legal, or tax advice. Any events, rewards, online promotions, or related information mentioned herein should not be considered a recommendation, solicitation, or invitation to purchase, sell, trade, or otherwise deal in any crypto assets. Crypto assets are highly volatile and may result in loss. The availability of WEEX services, products, and related events may vary by region. You are responsible for ensuring that your participation is in accordance with applicable local laws and regulations.