Grok chats leaked online: 370,000 private conversations exposed
The Grok chatbot, developed by Elon Musk's artificial intelligence company xAI, has come to the fore after users' private conversations were leaked online. It is reported that approximately 370,000 conversations have become accessible via search engines.
12punto
The fact that private conversations held by users on the Grok platform have become publicly accessible through search engines has raised serious concerns regarding data privacy. It has emerged that hundreds of thousands of dialogues users conducted with Grok have been indexed and are accessible on the internet.
USER DATA AT RISK
Grok's "share" feature results in the creation of public links instead of sending chats only to the intended recipient. This situation has led to the dissemination of users' private information without their knowledge. As a result of searches conducted on the internet, it was determined that approximately three hundred thousand Grok conversations were accessed within a single day. According to some sources, this number has exceeded three hundred and seventy thousand.
ILLEGAL CONTENT ALSO PRESENT
Among the leaked conversations, in addition to personal content such as secure password requests, nutritional advice, and health consultations, there is also illegal content. For instance, it was detected that in some conversations, the user was provided with information on the production of illegal drugs in a laboratory setting.
EXPERTS HIGHLIGHT PRIVACY RISKS
Prof. Luc Rocher from the Oxford Internet Institute stated that such systems could expose critical user data such as identity, location, and mental health, noting that this could be a "privacy disaster." Rocher emphasized that information that ends up on the internet becomes permanent.
AI SECURITY BACK ON THE AGENDA
This incident is not the first example of privacy issues with AI chatbots. Similar problems have occurred previously on platforms such as ChatGPT and Meta AI. The company xAI has not yet made an official statement. The incident has once again demonstrated that AI tools need to be re-evaluated in terms of privacy and ethical standards.