OpenAI Pulls Searchable ChatGPT Feature Over Privacy Concerns

OpenAI had earlier introduced a new feature which was to allowed users to make private conversations with ChatGPT searchable however this new feature was rolled back due to some reasons known to them. According to the OpenAI’s Chief Information Security Officer Dane Stuckey said on Thursday in a social media post stating that the team just removed a feature from @ChatGPTapp that allowed users to make their conversations discoverable by search engines, such as Google,” OpenAI’s chief information security officer Dane Stuckey said on Thursday in a social media post.

He further said that the business deleted the opt-in function, which he defined as  with the ‘’a short-lived experiment to help people discover useful conversations,” in part due to OpenAI’s emphasis on “security and privacy.”

“Ultimately we think this feature which first introduced had too many opportunities for folks to accidentally share things they didn’t intend to, so we’re removing the option,” Stuckey’s post added. Additionally, we’re trying to get the appropriate search engines to stop indexing information. All users will be able to access this change starting tomorrow morning.

Fast Company was the first to reveal on Wednesday that Google was indexing certain ChatGPT discussions. Shortly after newsletter writer Luiza Jarovsky announced on X on Thursday that private chats with ChatGPT were being made public, the reversion was implemented. According to Jarovsky, users were exposing their conversations to Google indexing when they used the chatbot’s sharing feature.

In order to share their discussions, users had to select a box to “make this chat discoverable,” which meant that their conversations would “be shown in web searches.” Public chats were anonymised to lessen the possibility that a user’s interactions with the bot may be used to identify them personally.

However, comments from Jarovsky’s tweets on X raised concerns that some people would unintentionally tick the “share” box without reading the fine print, leaving themselves vulnerable to disclosing private or humiliating information.

Jarovsky discovered instances of people using ChatGPT to talk about their anxieties in her first X post. More intimate examples abound in her comments, ranging from conversations about harassment to improvised therapy sessions.

Business Insider’s requests for comment were not immediately answered by Jarovsky or OpenAI representatives.

Although the business admits that there were unexpected and unforseen privacy compromises, the discoverability option was designed to encourage users to reuse outputs created in ChatGPT.

Although OpenAI provided clear visibility controls, it’s possible that many individuals were unaware of the consequences of turning on search indexing.

This serves as a reminder to use caution when entering data into AI chatbots. Features like sharing, logging, or model training might open the door for content to be made public even though a discussion is initially private.

In addition to reevaluating ChatGPT’s public sharing features, OpenAI claims to have being collaborating with Google and other search engines to eliminate indexed shared links for users.

Through the ChatGPT Shared Links dashboard, you can review your visibility settings and remove shared links if you have already shared a ChatGPT conversation.

By default, only those who had the URL could access shared ChatGPT links. However, users have the option to enable discoverability, which would enable search engines like Google to index the discussion.

Now that the setting has been removed and chats that have already been shared will no longer be indexed. OpenAI warns that caching may cause previously indexed content to momentarily still show up in search results.

Most importantly, the public share link and search engine visibility are not erased when you remove a chat from your ChatGPT history.