A Firebase Misconfiguration Exposed 300 Million AI Chat Messages
A researcher accessed 300 million messages from 25 million users of the AI chat app Chat & Ask AI through an exposed, unauthenticated database — a Firebase mistake found in 103 of 200 scanned iOS apps.
An independent security researcher found a major data breach affecting Chat & Ask AI, one of the more popular AI chat apps on Google Play and the Apple App Store, with more than 50 million users.
The researcher claims to have accessed 300 million messages from over 25 million users through an exposed database. Reported contents included discussions of illegal activity and requests for suicide assistance, among other sensitive material.
Chat & Ask AI is a "wrapper" app that plugs into multiple large language models — OpenAI's ChatGPT, Anthropic's Claude, and Google's Gemini — letting users pick which one they talk to. The exposed data included user files with complete chat history, which model was used, and other account settings. The exposure also reached data belonging to users of other apps built by Codeway, the app's developer.
The root cause was a well-documented category of mistake: a Firebase misconfiguration. Firebase is Google's cloud backend-as-a-service platform for building and scaling mobile and web apps. The most common version of the error is leaving Firebase's Security Rules set to public, which lets anyone who has the project URL read, modify or delete data without authentication.
That led the researcher, who goes by Harry, to build a tool that automatically scans apps on both Google Play and the Apple App Store for the same vulnerability. The results were not encouraging: 103 of 200 iOS apps scanned had the issue, collectively exposing tens of millions of stored files.
Source: Malwarebytes