A surprising privacy issue recently came to light when thousands of shared ChatGPT chats were discovered through Google Search. These chats, many containing deeply personal stories, were never meant to be public yet they ended up just a search away. The cause: ChatGPT’s “Share” feature The root of the issue lies in ChatGPT’s built-in “Share” option. When users share a chat, it creates a public URL that can be opened by anyone. What many didn’t know is that Google can also crawl and index these shared links, making them searchable on the web. A simple Google search using site:chatgpt.com/share revealed over 4,500 chats, some discussing mental health, trauma, addiction, relationships, work problems, and more. These conversations, though not automatically linked to a user’s name, can still contain self-disclosed personal or professional details that make them identifiable. Risks: More than just embarrassment What you should do right now If you’ve ever shared a ChatGPT conversation, take these steps immediately: OpenAI responds quickly OpenAI’s Chief Information Security Officer, Dane Stuckey, confirmed that the sharing feature had been removed, calling it a “short-lived experiment.” He admitted the tool, intended to help people find useful conversations, had backfired by making it too easy to share sensitive content accidentally. OpenAI is now working with Google and other search engines to remove already indexed conversations, but some may still appear temporarily due to caching. This situation is a strong reminder: AI chats feel private but they might not be. Always think twice before sharing, especially when the internet is watching.
