Shared Claude chats were blocked with robots.txt but never tagged noindex
Per WIRED, Anthropic blocked link-shared Claude chat pages in robots.txt but left off the noindex meta tag. Robots.txt asks crawlers not to fetch a page; it does nothing to keep that page out of a search index.
Reddit users had already been pulling up those conversations in Google with a site:claude.ai/share query. What turned up included a real patient’s medical report, documents with the names and phone numbers of elementary school students, and internal company files.
Up next
Shared Claude chats turned up in Google search results
Google was surfacing shared Claude chats: a site:claude.ai/share search returned a list of them, TechCrunch reports. Futurism found a real patient’s medical report among the results, plus files with elementary schoolers’ names and phone numbers.
Related stories
- Anthropic joins a shared robots.txt opt-out for AI training
- Apiiro finds Claude Code secrets in 1,122 GitHub files
- Email leak stopgap in Claude Code checks only shell calls
- Claude Code agent sat on a Pentagon network for two months
- Irregular CEO blames human error for AI agent breaches
- FlashProve proofs cost ~70-95x the GPU time of inference
Comments
No comments yet. Be the first.
Join the conversation
Sign in to leave a comment. It appears after moderation.