Claude shared chats briefly indexed by search engines

Search engines briefly indexed thousands of shared Claude conversations, exposing them to public discovery. Users on Reddit found the conversations by performing a simple Google search using the query site:claude.ai/share, which returned results for Anthropic's chatbot shared chats. The vulnerability stemmed from Anthropic's "Share with link" feature, which apparently lacked a noindex tag—an HTML directive that explicitly tells search engines not to include a page in their results.

Some of the publicly accessible shared chats reportedly contained sensitive information including crypto keys and legal queries. These discoveries were particularly alarming because users had likely believed their shared conversations were private and accessible only to those who received the specific share link. Anthropic responded to the issue quickly, and the Google search results disappeared soon after the problem was identified. However, the shared chats remained visible on Bing and Brave Search for a longer period, suggesting the remediation was not instantaneous across all search platforms.

Anthropics is not the first AI company to make this mistake. OpenAI faced an identical problem last year with its own "Share with link" feature and later chose to remove the feature entirely. In contrast, Anthropic has kept the feature active and addressed the indexing problem. Users who are concerned about their previously shared conversations can review and manage them by navigating to Settings → Privacy → Shared Chats, where they can assess what may have been exposed and make adjustments to their privacy settings.

The incident reveals a common security oversight in AI platforms: the "Share with link" feature in Claude lacked a noindex HTML tag that instructs search engines not to include pages in their results. This technical gap allowed search engines to index what users reasonably expected to be controlled sharing—a private link sent only to intended recipients. The exposure was particularly concerning because some shared chats contained sensitive data like cryptocurrency keys and legal queries, information users would never knowingly publish to the public web.

Anthropics's response was relatively swift on Google, where results disappeared soon after the issue was identified. However, the chats remained visible on Bing and Brave Search longer, suggesting inconsistent remediation across search platforms. OpenAI encountered the same problem last year with its own share feature and ultimately removed the feature entirely—a more drastic measure than Anthropic has taken so far. The incident underscores the need for AI companies to implement privacy safeguards as standard practice during feature development, particularly when users are given the impression that their data is restricted.

No comments yet. Be the first to share your thoughts!