OpenAI Removes Public Chat Sharing Feature After Private Conversations Appear In Search Results

OpenAI has removed ChatGPT shared chats from Google search after privacy concerns, ending a feature that made personal conversations publicly accessible through indexed links.

ChatGPT Removes Option to Have Private Chats.

Source : Unsplash/Jonathan Kemper

OpenAI has quietly removed the option that allowed ChatGPT users to make their shared conversations discoverable through Google search results—effectively wiping those links from public view. A quick search using site:chatgpt.com/share on Google now returns no results.

The change follows growing concerns about privacy, after users unknowingly made their private ChatGPT chats publicly accessible on search engines. Although the links are still turning up on Bing and DuckDuckGo, OpenAI says that's temporary. “The feature is still rolling out to all users,” explained Chief Information Security Officer Dan Stuckey. “Ultimately, we think this feature introduced too many opportunities for folks to accidentally share things they didn’t intend to, so we’re removing the option.”

Stuckey added that OpenAI is actively working to remove any remaining indexed content from other search engines.

An OpenAI spokesperson confirmed to PCMag that the public chat sharing option was part of a broader “experiment”—one that has now come to an end.

How Private Chats Ended Up in Public Search Results

When OpenAI launched the sharing feature in May 2023, the goal was to simplify the way users distributed their ChatGPT conversations—moving away from screenshot sharing to a cleaner, link-based format. But the feature came with an optional checkbox labeled “Make this chat discoverable,” allowing those links to be indexed by search engines.

Despite the checkbox being optional, some users either misunderstood its implications or overlooked the fine print that read, “Allows it to be shown in web searches.” As a result, sensitive and personal chats began surfacing in search results.

And what surfaced was deeply personal. In some indexed chats, users asked for support dealing with loneliness and trauma. Others sought help navigating toxic relationships or even explored conspiracy theories. One person, for example, asked the chatbot whether Cape Cod, Massachusetts, was “the deep state” complete with “secret tunnels” and “giant generators.”

Privacy, Discoverability, and Digital Footprints

While shared links did not expose user names or account details, they often contained enough context to make identities guessable—especially if a user mentioned names or specific personal details within the conversation.

A Google spokesperson clarified that the indexing wasn't initiated by search engines. Rather, OpenAI enabled the visibility. For users concerned about their privacy, the good news is that shared links can be edited or deleted. And if you delete your ChatGPT account entirely, your shared links are removed as well.

Still, not everyone is alarmed. Some users may have intentionally made their chats public—using alternate accounts or hoping to connect with others online. But for most, the feature seems to have created confusion and unintended exposure.

“People talk about the most personal shit in their lives to ChatGPT,” OpenAI CEO Sam Altman said in a recent interview, according to TechCrunch. “Young people especially use it as a therapist, a life coach... asking, ‘What should I do?’”

Adding to privacy concerns, OpenAI is also legally obligated to retain and potentially disclose user conversations in the event of legal proceedings.

SEO Community Saw a Goldmine

While many viewed the discoverable chats as a privacy misstep, others in the digital marketing world saw opportunity. On Reddit, one SEO professional called the indexed conversations a “goldmine.”

“These conversations reveal exactly what your audience struggles with,” they wrote. “The questions they’re too embarrassed to ask publicly.”

By studying real user queries to ChatGPT, marketers believed they could gain a direct line into customer needs—something that previously required extensive keyword research and interviews.

Not the First Time This Has Happened

OpenAI isn’t the only tech giant to face pushback over inadvertently public AI chats. Meta ran into a similar issue in June when users discovered that shared conversations in its AI app were appearing in other people’s feeds. The company responded by adding a warning prompt before making any chats public.

Google also had its own stumble. After rebranding Bard to Gemini in early 2024, users found that Gemini chats were showing up in search results using the query site:gemini.google.com/share. Google quickly stopped indexing those pages.

Interestingly, some users suspect there’s more crossover between Google and OpenAI than meets the eye. On Reddit, people have noted that ChatGPT’s responses often mirror Google’s AI overviews—and sometimes pull directly from search engine results.