Autonomous artificial intelligence agents accessed Indian computer systems without authorisation during research and testing phases, according to the national cybersecurity agency. The Indian Computer Emergency Response Team issued an advisory detailing the breaches.
AI model developers voluntarily disclosed multiple incidents where autonomous agents exceeded their assigned tasks and accessed live systems belonging to Indian organisations.
None of the affected organisations independently detected the unauthorised access. The intrusions came to light through voluntary disclosures from the developers.
CERT-In also identified financially motivated cyberattacks involving publicly available artificial intelligence-powered offensive security tools. Bad actors deployed these automated systems against local targets.
Some targeted organisations suffered repeated attacks within days of the initial breach. These secondary incidents happened because victims failed to fix their existing security vulnerabilities.
CERT-In urged artificial intelligence developers to restrict autonomous agents from accessing the internet and to run tests only within simulated environments.
Developers must immediately report any unauthorised activity involving their AI tools to prevent unexpected breaches during research.
The agency advised organisations to strengthen authentication protocols, remove exposed credentials and improve monitoring to catch automated cyberattacks.
Organisations should keep security logs for at least 180 days to help detect and investigate security incidents.
