Exposed! OpenAI Removes Controversial ChatGPT Feature After Privacy Backlash
OpenAI turned off a feature in GPT-3 that let the AI make answers to users’ questions that could be indexed by Google Search and found there. This is because more and more people are worried that it could let people break the privacy in conversations without them knowing. An official statement from Chief Information Security Officer (CISO) Dane Stuckey on X (formerly Twitter) made it clear that the choice has been made and will not be changed.
We just removed a feature from @ChatGPTapp that allowed users to make their conversations discoverable by search engines, such as Google. This was a short-lived experiment to help people discover useful conversations. This feature required users to opt-in, first by picking a chat… pic.twitter.com/mGI3lF05Ua
— DANΞ (@cryps1s) July 31, 2025
When the tool came out earlier this year, users could choose whether to have some of their chats shown in a public search. But it got a lot of bad feedback when some users found private, sensitive conversations in the search results that were open to everyone.
Indexing Led to Accidental Exposure of Private Conversations
Users had to click on a chat and then confirm they wanted to share it with their friends, so it was an opt-in feature. However, many users seem to have had the wrong idea about what they were signing up for. A study by Fast Company found that more than 4,500 conversations were linked on Google. Some of these conversations included names and locations of different users.
OpenAI CEO Sam Altman had earlier acknowledged the profoundly personal nature of many ChatGPT conversations, stating, “People talk about the most personal shit in their lives to ChatGPT.”
Deletion Not Enough to Reverse Exposure
Critics also say that even if a user deletes or turns off group chat, it may still be logged by search engines. The logs stay there until they are deleted by hand or until the algorithmic logs are refreshed, whichever comes first. OpenAI has said that it is already working with search engines to find creative ways to take down information that has already been shared. They say this is part of their goal.
In a world where AI can talk, how safe is it that our privacy will be?
