Claude shared chats and Artifacts exposed on Google

Over the weekend, Reddit users discovered that shared Claude chats and Artifacts—interactive mini apps built inside Anthropic's AI assistant—were publicly searchable on Google using operators like "site:claude.ai/share". The exposed content included health records, private company documents, names and phone numbers of children, clinical trial results, internal company documents, employee reviews with personal information, and code. In one case, a chat labeled "shared by Anthropic" showed Claude producing erotica, violating its usage policy. Anthropic blamed users, stating that share links only appear in search results when posted publicly on forums or social media, and that private links remain unindexed. The company does not share chat directories or sitemaps with search engines. Google spokesperson Ned Adriance noted that Google does not control what pages are made public, and site owners can use directives to prevent crawling. As of Monday afternoon, a test search no longer returned results, suggesting remediation. A similar issue occurred last year when hundreds of Claude chats were indexed; Google estimated under 600 conversations. Also last year, a researcher scraped around 100,000 public ChatGPT conversations. Users can review their shared chats via Settings > Privacy > Shared Chats.