More than 300,000 private Grok chats leaked by Google

Hundreds of thousands of private chats with Grok have been indexed by Google, revealing a serious data breach risk just weeks after a similar incident with ChatGPT.

A shocking discovery shows that more than 300,000 private chats with Grok are publicly available on Google.

The reason is the Share feature, which creates a link but does not limit access.

This causes search engines to automatically crawl and index private content, making it public.

Many users have shared sensitive information such as emails, phone numbers and health data.

The incident poses a serious risk of information misuse and personal reputation damage.

What's worrying is that Grok doesn't explicitly warn users about the risks of pressing the Share button.

Security experts say this is a systemic bug and shows a vulnerability across the AI ​​industry.

Without better data protection, AI platforms will lose trust — their most valuable asset.

Dear readers, please watch more videos : AI Trash - New problem on social networks VTV24

Source: https://khoahocdoisong.vn/hon-300000-doan-chat-grok-rieng-tu-bi-google-lo-post2149047545.html