Home / Technology / Your AI Bot Conversations Aren't Private
Your AI Bot Conversations Aren't Private
27 Jul
Summary
- Chatbot conversations, including sensitive data, can become publicly searchable.
- Users unknowingly share personal medical, financial, and student data via AI.
- AI platforms warn of public sharing, but users miss web crawler risks.
Conversations with AI chatbots, including sensitive personal information, have been discovered to be publicly searchable on platforms like Google. Users have reported finding medical histories, internal business documents, and even student directories exposed. This occurs when users utilize the "share" feature for their AI-generated content, such as code or documents, inadvertently making it accessible via a public link.
While AI providers like Anthropic offer a warning that shared conversations are public, users may not realize these links can be indexed by web crawlers and appear in search results. This exposure risk is not unique to one AI model, with similar incidents reported for other chatbots. Anthropic states that shared links only appear in search if posted publicly online, and private links remain confidential unless further shared.
Users concerned about shared data can manage their "shared chats" and "shared artifacts" through their account settings to revoke access. However, content already indexed by search engines may remain discoverable for some time. It is generally advised to avoid sharing sensitive information with AI chatbots and to be cautious with the "share" functionality.
