Hundreds of private Claude conversations, including crypto credentials and legal details, were indexed by Google due to a missing noindex tag.
A Reddit user discovered on July 25 that Google had indexed hundreds of shared Claude AI conversations, exposing sensitive data. The breach included crypto wallet seed phrases, Social Security numbers, and legal discussions, all accessible via a simple search query.
The issue stemmed from a missing noindex meta tag on shared links, despite Anthropic’s existing robots.txt file. Shared links labeled “Anyone with the link” were publicly accessible, allowing Google to index them. The leak revealed personal, financial, and professional details users had shared with the AI.
The Reddit thread documenting the findings received over 4,000 upvotes, highlighting the scale of the exposure. No immediate market reaction was reported, but the incident raises concerns about AI privacy and data security.