Private Claude AI Chats Surface in Google and Bing Search Results

Sensitive and personal conversations with Anthropic's AI chatbot Claude were unexpectedly found to be publicly accessible and indexed by major search engines, including Google and Bing, over the weekend. The issue, first reported on Reddit, revealed instances of users seeking advice on political affiliations, attorney ethics in Kansas, and engaging in erotic role-play. These chats became discoverable due to Claude's feature allowing users to share specific conversation threads via a public URL.

Anthropic, the developer of Claude, employs a standard web protocol called robots.txt to instruct search engine crawlers, such as those used by Google and Bing, not to index content designated as private or shared. According to a snapshot from the Wayback Machine, Anthropic's robots.txt file has directed web scrapers away from shared chats since at least September 2025. However, preventing pages from appearing in search engine results can be more complex than simply using robots.txt.

Privacy implications and technical workarounds

While Bing still displays numerous results when searching for "site:claude.ai/share," indicating it may not fully adhere to robots.txt directives in all cases, its own documentation suggests that developers should also implement a "noindex" HTML tag on individual pages for robust protection. Some of the problematic chats highlighted in the initial Reddit post were no longer available by the time they were reviewed, and similar search queries that previously worked on Bing no longer yield results on Google. Google's developer guidelines state that the company may ignore robots.txt instructions if a page is linked to from elsewhere and lacks a "noindex" tag in its header or response. A review of several exposed Claude chat pages found they did not contain this crucial "noindex" tag.

Google spokesperson Ned Adriance clarified that the indexing of these shared Claude chats is the responsibility of Anthropic, stating, "Neither Google nor any other search engine controls what pages are 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." Anthropic has not yet commented on why the "noindex" tag was omitted from the shared chat pages. This is not the first time Anthropic has faced scrutiny over this issue; a similar incident last September led the company to reiterate its reliance on robots.txt, a method that, as a Forbes report noted, offers no guarantee against indexing by search engines.

Despite the challenges in ensuring complete index prevention, AI labs continue to utilize robots.txt for other critical purposes, such as preventing their websites from being used for AI training data. Many platforms promise creators that their content will not be scraped for training if they disallow specific crawlers in their robots.txt files, a practice that AI developers themselves also follow.