TechCrunchAI & LLMs

Google Leak Exposes Thousands of Claude Chats Containing Medical Records and Private Data

Over the weekend, a privacy lapse left countless Claude conversations and Artifacts—interactive mini-apps built within the AI assistant—publicly searchable on Google. Reddit users discovered that typing "site:claude.ai/share" into the search engine returned a long list of shared links. Some of those chats reportedly contained sensitive health records, private company documents, and the names and phone numbers of children. The incident underscores how easily data can slip outside intended sharing circles when tools designed for small-group collaboration meet the open web.

The root cause traces back to Claude’s “share chat” feature, which generates URLs that let anyone with the link view a conversation. Anthropic’s interface warns, “Anyone with the link can view,” implying the feature is meant for friends or colleagues—not the entire internet. But when users post those links on public forums or social media, search engines like Google can index them. Anthropic told TechCrunch that share links only appear in search results if they’ve been shared publicly—a link sent privately stays hidden from search. The company added that it does not submit chat directories or sitemaps to search engines, and that these links are not guessable. Google spokesperson Ned Adriance echoed that point, stating that Google respects site owners’ directives and does not control what pages are made public; the pages in question were indexed across multiple search engines.

Before the exposure was remediated—a test search by TechCrunch on Monday afternoon returned no results—Futurism found a detailed medical report, clinical trial results with patient names, documents listing schoolchildren’s names and phone numbers, internal company memos, and employee reviews containing personal information. Exposed Artifacts included code and work notes. In one instance flagged by Fortune, a chat labeled “shared by Anthropic” depicted Claude generating erotica, a violation of the company’s usage policy. It’s unclear how that content was produced. The incident mirrors a similar 2023 event where hundreds of Claude chats were indexed by Google, and a separate researcher scraping roughly 100,000 publicly shared ChatGPT conversations. Users can review which of their Claude chats have public links by navigating to Settings > Privacy > Shared Chats.

Source: TechCrunch

← Back to News