Hundreds of conversations between users and Anthropic's AI chatbot, Claude, were discovered to be publicly accessible through search engines such as Google, Bing, and DuckDuckGo. These chat logs, some containing sensitive personal and work-related information, appeared in search results when specific site-based queries were made.
The issue arose because users who chose to share their chat links inadvertently made the conversations discoverable online. Although the chat interface warns that "anyone with the link" can view the content, it does not explicitly inform users that these links might be indexed by search engines.
Users on Reddit initially uncovered over 200 shared conversations spanning at least 25 pages of search results. The chats covered a wide range of topics, including personal queries, professional matters, and even proprietary research. Examples included a user asking Claude to draft an unpublished blog post on cloud security with corporate details, and another seeking help with resume content containing personal identifiers.
Anthropic responded by removing the indexed links from search engine results over the weekend. A company spokeswoman emphasized that users retain control over when to share their conversations and that links are not "guessable or discoverable" unless deliberately shared. However, once shared, the content becomes publicly accessible and may be archived by third parties.
This incident echoes similar past events involving AI chatbots. Last year, OpenAI’s ChatGPT faced a comparable problem with chat logs being publicly accessible, prompting changes to access controls. Likewise, Grok, Elon Musk’s AI chatbot on X, experienced widespread exposure of chat logs through online searches.
A Google spokesperson clarified that the company does not control web content publication but provides website owners with tools to manage indexing. The removal of Claude chat logs from search results likely involved Anthropic using such tools to block links from being crawled and indexed.
