Shared conversations from Anthropic’s Claude AI assistant were publicly discoverable via Google search over the weekend, after Reddit users found that a simple search operator could surface a long list of chats and Artifacts – the interactive documents and mini-apps users can build within the platform.
The exposed content reportedly included health records, clinical trial data, the names and phone numbers of primary school-aged children, internal company documents and employee reviews containing personal information. Exposed Artifacts included code and work notes. At least one shared chat attributed to Anthropic was found to contain sexually explicit content – material that Anthropic’s usage policy explicitly prohibits.
The issue traces back to Claude’s “share chat” feature, which generates a URL that allows anyone with the link to view a conversation. The feature’s own interface warns that “anyone with the link can view” – language users have reasonably read as sharing with small, trusted groups rather than broadcasting to the open internet. A comparable sharing function in Google Docs does not result in documents becoming indexable by search engines.
Anthropic’s position is that it does not supply chat directories or sitemaps to search engines, and that shared links are neither guessable nor discoverable unless a user posts them somewhere publicly accessible, such as a forum or social media platform. Spokeswoman Amie Rotherham stated that a link sent privately stays out of search results. The company, in effect, attributed the exposure to user behaviour rather than a platform fault.
Google spokesperson Ned Adriance said the company does not control which pages are made public on the web, and that site owners have clear tools available to prevent crawling or indexing.
The issue was first flagged on Reddit on Saturday and reported by 404 Media on Monday morning. By Monday afternoon, the same search method no longer returned results, suggesting the exposure had been resolved – though neither Anthropic nor Google has confirmed the precise remediation steps taken.
This is not the first time Claude conversations have surfaced publicly at scale. Last year, Google estimated it had indexed just under 600 shared Claude chats before those pages disappeared from search results. A separate 2025 incident saw a researcher scrape approximately 100,000 publicly shared ChatGPT conversations, underlining that link-based sharing features across AI platforms carry real privacy risks when users remain unaware of how far their content can travel.
Users can review and manage their shared Claude conversations by navigating to Settings, then Privacy, then Shared Chats.
