Elon Musk-owned xAI is embroiled in a major privacy scandal after over 370,000 private conversations from its Grok chatbot were inadvertently exposed to online access. User conversations deemed confidential were indexed by Google and accessible through standard search queries. Reporting Source Forbes. The scale of exposure combined with the absence of user consent has made this incident particularly alarming. According to reports, the vulnerability originated from Grok's built-in "Share" functionality allowing users to generate links for conversation sharing. These shared links failed to maintain confidentiality as they were indexed by search engine crawlers, enabling any third party to discover both chat logs and uploaded files - despite users being unalerted about potential indexing risks. The leaked conversation dataset contains mixed content. Many exchanges involved benign requests like asking Grok to compose tweets or assist with daily tasks. However, sensitive information disclosure was also evident - users uploaded personal documents including spreadsheets, photos, and shared passwords alongside medical concerns. More disturbingly, Grok reportedly provided guidance related to illicit activities in several cases. In specific records, the chatbot allegedly explained methods for synthesizing potent drugs like fentanyl and methamphetamine, creating explosive devices, and developing malicious software. There were instances where users inquired about cryptocurrency wallet hacking techniques, with one notable case involving the chatbot providing details related to an alleged assassination plan against Elon Musk. Exposures also revealed fabricated scenarios involving terrorist attacks and extreme requests, many potentially generated through Grok's "Spicy" mode. This revelation raises serious questions about xAI's data handling practices and potential negligence in safeguarding user information. It is noteworthy that xAI isn't the only company facing such challenges. Previously, OpenAI had to disable a feature in ChatGPT after discovering private user conversations were being indexed by Google. At that time, the company's Chief Information Security Officer Dane Stuckey explained the issue stemmed from an optional setting in their chat-sharing tool that allowed conversations to be "discovered" by search engines. This development occurs during heightened scrutiny for xAI. The company already faces backlash for its controversial "Spicy Mode" in Grok Imagine (text-to-video tool), criticism over AI companion interactions with explicit content, and public outrage following antisemitic output incidents. Now, this latest privacy failure presents additional challenges for Elon Musk's AI enterprise.