ChatGPT conversations once meant for personal use are now appearing in public search results.
A series of startling discoveries have revealed that some conversations shared via ChatGPT’s “Share” feature have been indexed by Google, putting sensitive user information in the public domain.
Originally designed to promote user collaboration, shared ChatGPT links are now turning up in simple Google searches. To make matters worse, some of them even contain legal names, email addresses and even private emotional confessions.
OpenAI’s sharing feature lets users generate unique URLs linking to their AI conversations. However, when these links are posted online, or if users opt to make them “discoverable,” they can be picked up by search engines. Many users, it seems, didn’t expect that their chats, once thought to be semi-private, would become easily searchable.
Numerous online communities, such as Reddit, have begun cataloging examples of private conversations exposed in the public domain. Some of the most shocking ones include full resumes, children’s names, and conversations detailing mental health struggles.
In some cases, people even stumbled across discussions that appeared to involve illegal activity or personal business transactions. The fallout has sparked a series of ethical questions. One user recounted warning a sex worker after finding her identity exposed via a shared chat, only to meet backlash. The ease of access to such data raises serious concerns about doxxing, harassment, blackmail, and reputational harm, especially when conversations touch on controversial or intimate topics.
In response to the backlash, OpenAI swiftly removed the discoverability feature from ChatGPT. Chief Information Security Officer Dane Stuckey confirmed the change via X, calling it a “short-lived experiment” that unintentionally paved the way for severe privacy violations.
OpenAI emphasized that users had to opt in to make conversations searchable, but acknowledged the stakes were too high. The company is now taking steps to purge these indexed links from search engines entirely, aiming to limit further exposure.
If you’ve shared a ChatGPT conversation in the past and are concerned it may be publicly accessible, here’s how to protect your information:
1. Open ChatGPT and log into your account
2. Click your profile or go to Settings
3. Navigate to Data Controls > Shared Links
4. You’ll see a list of all your previously shared conversations
5. Click the three dots next to each and select Delete Link
Note: Deleting the chat itself from your ChatGPT history does not delete the shared version.
After deleting the shared link, the page will return a 404 error, but it may still appear in Google search results for a while. To request faster removal:
1. Visit Google’s Remove Outdated Content Tool
2. Click the New request button
3. Select a relevant option from the menu
4. Follow the on-screen steps to submit a removal request
These steps can significantly reduce the chance of your private content being publicly discoverable, even after OpenAI’s feature rollback.
Specialized software like Bitdefender Digital Identity Protection can help you keep a close eye on your online data, including traces from services you no longer use.
It constantly scans both the public and Dark Web for your data, notifies you if you’ve been compromised by a breach, and helps you patch vulnerabilities in your digital footprint quickly with simple, one-click action items.
tags
Vlad's love for technology and writing created rich soil for his interest in cybersecurity to sprout into a full-on passion. Before becoming a Security Analyst, he covered tech and security topics.
View all postsMay 16, 2025