ChatGPT Conversations Are Still Accessible Through a ‘Shady Technique’ Despite Attempt to Remove Them From the Internet; Grok Chats Have Also Surfaced Up – Wccftech
AI firms have breached user privacy by allowing individuals to access chats through Google Search, and it seems that attempts to resolve this have failed.
For those unaware, folks on the internet recently discovered that the public can easily access ChatGPT’s chats through a simple Google Search, which exposed over 100,000 AI conversations from different people. According to OpenAI, the firm is working towards de-indexing these chats, but they aren’t just on Google Search, as according to Digital Digging, several conversations are still accessible through Archive.org’s Wayback Machine, causing huge concerns not just for the people who have seen a privacy compromise, but for OpenAI as well.
When the Wayback Machine inquired whether OpenAI had contacted them to remove the respective Google links, their Director, Mark Graham, claimed that no individual from the AI firm had requested removal. Here’s what he had to say:
I can/will tell you we have not gotten, or honored any requests for (large scale) URL exclusion of “chatgpt.com/share” URLs. If OpenAI, the rights holder for material from the domain chatgpt.com, asked for the exclusion of URLs from the URL pattern chatgpt.com we would probably honor that request. However, they have not made such a request.
The Wayback Machine technically saves every publicly accessible page on the Internet through “snapshots” that have time-stamped versions, even if OpenAI deletes them from Google. Since there hasn’t been a request to take them down yet, individuals can still access them. This means that for those who know how to work with Archive, they can still look at chats. While not all of them are interesting, some out there were pretty wild, although we won’t mention them for the sake of user privacy.
At the time of writing, Grok’s chats are readily accessible on Google Search, so it is safe to say that this phenomenon is a wider one, occurring with mainstream models used by consumers. It does indicate about the privacy concerns associated with using AI models, and more importantly, these chats can also be used in legal matters as well, so you need to be a bit more careful on how you talk with chatbots over the internet.
Some posts on wccftech.com may contain affiliate links. We are a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for sites to earn advertising fees by advertising and linking to amazon.com
© 2025 WCCF TECH INC. 700 – 401 West Georgia Street, Vancouver, BC, Canada
