The AI startup xAI has inadvertently published hundreds of thousands of conversations between users and the Grok chatbot on Google and other search engines. In many cases, confidential information was disclosed without user consent, reports Forbes.
Each time a user clicks the “share” button in the chat, a unique link is generated. This link allows the conversation to be shared via messengers, email, and other means. It turns out these links are also indexed by search engines like Google, Bing, and DuckDuckGo.
In other words, clicking the “share” button makes the conversation publicly accessible to anyone on the internet.
According to the publication, more than 370,000 dialogues have been indexed. The published conversations range from simple business tasks like writing tweets to more questionable content, such as generating fake news about terrorist attacks in Kashmir or attempts to hack a crypto wallet.
Personal inquiries about medicine, psychology, and passwords can be found. Uploaded images, spreadsheets, and text documents were also made accessible.
Some requests contained obscenities and xenophobic remarks. Grok even provided instructions for making drugs, self-replicating malware, bombs, and advice on suicide methods. In one conversation, the chatbot offered a plan to assassinate Elon Musk.
Previously, a similar scandal occurred with OpenAI’s ChatGPT. The company then explained the indexing of conversations as a test feature and disabled it. Musk mocked the competitor, while Grok’s official account assured that the service has no such feature.
BriansNotion Unlike OpenAI’s ChatGPT, where shared chats can be indexed by Google, Grok has no such sharing feature. We prioritize privacy: private chats are deleted within 30 days, we don’t sell data, and advise against sharing sensitive info. See https://t.co/m7tSshVPLZ for…
— Grok (@grok) August 1, 2025
It is unknown when the indexing of conversations was activated. Users had complained about it as early as January.
Google Search have indexed insane number of Grok conversations but do those users know?? That their conversations with Grok can be seen by others via Google Search. pic.twitter.com/7GszakeRlB
— Gagan Ghotra (@gaganghotra_) January 20, 2025
It is likely that some requests about drugs and weapons were made by cybersecurity specialists as part of AI testing. However, even professional researchers were unaware that their dialogues were being indexed by search engines.
Google explained that resource owners themselves decide on the indexing of pages. Previously, the company allowed this with Bard chats but removed the feature in 2023.
Meanwhile, marketers have already begun using the published Grok conversations for SEO. Discussions on LinkedIn and BlackHatWorld explore how they can be used to promote brands in Google’s search results.
Back in July, Sam Altman stated that personal conversations with ChatGPT are not legally protected. If necessary, a court can request all the information users have shared with the chatbot.