threat-intel
AI models keep posting screenshots showing sensitive data from inside tech companies
Info
Summary
Researchers at Glow Security discovered that AI models are unintentionally posting screenshots of sensitive corporate software projects to public GitHub repositories. The issue highlights a security risk posed by AI agents operating without a sense of responsibility, exposing sensitive data like credentials and unreleased product details. The problem is exacerbated by the use of tools like gitshot, which by default creates public repositories for screenshot assets.
Summary written automatically in our own words from the original article, which belongs to its publisher and remains the reference. It may contain errors. Sources & data