Your Correspondence With Claude May Have Been Published on Google.
Chatbots are designed to make your conversations feel personal. They’re incredibly pleasant to interact with, overly friendly, and mimic the feeling of chatting with a friend online—assuming that said friend can instantly generate a whole wall of text in a matter of seconds. However, in reality, they’re far from personal, especially if your definition of “personal” includes “for your eyes only.”
Claude’s chat messages were visible on Google this weekend.
Some Claude users can confirm this fact after the weekend. As Cyber Security News reports , links to Anthropic’s Claude chats appeared in Google search results, indicating that search engines like Google have indexed users’ chats. Over the weekend , a viral Reddit post demonstrated how simply adding “site:claude.ai/share” to search queries directly found real Claude chats on Google.
If a user named Claude discussed the keyword you were searching for and Google indexed the conversation, you could find it and read the transcript. According to Cyber Security News, this included lawyers discussing legal strategies, engineers breaking down technical issues, and users sharing candid accounts of their personal struggles. Depending on the conversation, it could be anything from mildly awkward to downright humiliating.
Now, a few disclaimers: Anthropic quickly nipped the issue in the bud after a Reddit discussion sparked outrage. If you try to search for “site:claude.ai/share,” you’ll simply get an unsuccessful search result, so all your Claude chats should remain safe going forward. Perhaps most importantly, it appears these indexed chats didn’t affect all Claude users; rather, they were likely the “share links” Claude generates when you share a conversation with someone else. The glitch appears to be related to this feature—if you shared a chat from Claude, it could have ended up in this searchable database.
However, not all of the leaked conversations met these criteria. A similar situation occurred in September , and, as Futurism reports , at least one affected user denied sharing their conversations. If this claim is to be believed, it’s possible that some of Claude’s leaked conversations also belonged to users who never used the messaging feature.
AI-powered chats are not private.
Even if you’re confident that your conversations with Claude weren’t shared with third parties, or trust that the leak has been fixed, that doesn’t mean you should treat your conversations with your AI chatbot with the assumption of privacy. Even if you never shared any conversations with Claude and therefore likely weren’t affected by this issue, your conversations may not be stored in your account. While you may think what you say to your AI remains between you and the chatbot, everything you say to your bot is transmitted to Anthropic and used to train future models—unless you explicitly opt out.
This isn’t unique to Anthropic: most, if not all, AI companies operate this way, and each has its own set of rules. For example, in some cases, your chats may be directly viewable by humans—something you might not consider when asking your AI bot for help with a personal issue.
Even if you opt out of data collection, the company may still retain your chats for a certain period of time. For example, Anthropic will retain your chat data for 30 days after you delete a conversation. So, not only is this data stored on Anthropic’s servers, but it also remains there after you can no longer access it. Furthermore, Anthropic isn’t the only company facing indexing issues: while this is the second time Claude users have experienced this, ChatGPT users encountered a similar issue last August .
If you do decide to use chatbots, never share personal or sensitive information with them —especially sensitive data like health information . Artificial intelligence is undoubtedly here to stay, and many find it useful, but if you care about your privacy, these tools simply don’t have your best interests in mind.