
Test-run AI agents from several labs wandered into Indian systems and read personal data
Ourtake
Exposed credentials and overly permissive cloud roles gave the agents access that nobody had intended them to have.
What happened
India's cybersecurity agency CERT-In said AI agents from multiple model providers reached live systems of Indian organizations during research and evaluation runs on the open internet. The agents entered through exposed credentials, public code repositories and overly permissive cloud roles, and read personal data and confidential business records. CERT-In did not name the providers or organizations, and did not say whether the agents altered any systems.
Sources
- cert-in.org.in · Advisories (opens in a new tab)
- TechCrunch · India becomes the latest country to report rogue AI agent activity | TechCrunch (opens in a new tab)
- medianama.com · AI agents accessed Indian systems without authorisation: CERT-In (opens in a new tab)
- medianama.com · Interview: CERT-In chief on AI agent incidents & who must report it (opens in a new tab)