ChatGPT user conversations exposed via Google Search indexing
Researchers discovered that fragments of private ChatGPT conversations were being indexed and exposed in Google Search results through a bug in the web browsing feature. OpenAI confirmed the unintended data exposure affected a limited number of users and stated it was fixed upon discovery.
What happened
Researchers Jason Baker and Slobodan ManiΔ discovered that fragments of ChatGPT user conversations were being indexed by Google Search and appearing in search results when users employed the web browsing feature. The issue involved a web browsing mode that appended user query portions to URLs with an internal parameter labeled "hints=search", making these fragments visible and indexable by Google. OpenAI confirmed the problem affected a limited number of users and stated the flaw was fixed upon discovery.
Reported impact
The bug exposed private conversation snippets to public visibility through Google Search results, though OpenAI stated the leak did not include passwords or directly sensitive personal data. The incident occurred without user action to share conversations publicly, distinguishing it from earlier incidents where shared chat links were indexed.
Index analysis
The leak resulted from a technical integration issue between ChatGPT's web browsing functionality and Google's indexing practices. The vulnerability involved the web search parameter passing portions of user input through URL parameters that Google automatically indexed.
Classification
Sources and evidence
Sources not yet published
This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.