Claude chats appeared in Google and Bing results

Some Anthropic Claude conversations became searchable on Google and Bing over the weekend, exposing users' questions about political parties, Kansas legal-ethics rules, and erotic role-play. The issue appears to have been first flagged by a Reddit user.

Claude lets users create public URLs for “snapshots” of individual chats. Anthropic has used a robots.txt file to tell web crawlers not to index those shared conversations since at least September 2025, according to the Wayback Machine. That instruction, however, does not necessarily prevent pages from appearing in search results.

Why shared Claude chats were indexed

Bing’s documentation recommends that site owners use both robots.txt and a noindex tag on individual pages. Google similarly says it may ignore robots.txt when a page is linked from elsewhere and lacks either a noindex HTML tag or an x-robots-tag response header.

WIRED reviewed a sample of the exposed Claude pages and found that they did not contain the noindex tag recommended by both search engines. Bing still displayed “about 612 results” for the query site:claude.ai/share at the time of publication, while the results no longer appeared in Google for the same search. Some pages identified in the Reddit post had also been deleted.

Google spokesperson Ned Adriance said the issue was Anthropic’s responsibility:

Recommended reading

AI recording tools capture conversations without warning

“Neither Google nor any other search engine controls what pages can be made public on the web, and these pages were indexed across many search engines. We give site owners clear controls to decide whether pages can be crawled or indexed, and we always respect those directives.”

Microsoft, which owns Bing, did not comment before publication. Anthropic did not respond to multiple requests for comment or explain why its shared-chat pages lacked noindex tags.

Robots.txt is also used for AI crawlers

Anthropic faced similar criticism last September and told Forbes that it relies on robots.txt to prevent crawlers from accessing shared chats. The same mechanism is also central to agreements in which AI labs promise not to use creators' websites for training when those sites disallow certain crawlers.

Anthropic, Meta, and OpenAI each include disallow instructions in their chatbot sites' robots.txt files, blocking competitors' crawlers from accessing those areas. OpenAI and Meta did not comment, while Google did not address questions about competitors blocking its Google-Extended training crawler.

Although shared Claude chats have stopped appearing in Google results, WIRED found that the reviewed pages still lacked noindex tags. They could therefore become searchable again. Claude users can manage access under Settings > Privacy > Shared chats.