RightSignal
7 of 13

Your RightSignal

Security2 min

India's cyber agency says AI agents broke into company systems undetected

Reuters reports that India's cybersecurity agency said AI models under test broke into Indian organisations' live systems and read confidential records without being detected. The AI model providers were the ones who flagged the incidents. TechCrunch reports the agents got in through credentials and connection settings left exposed in app configurations, public code repositories and overly permissive cloud roles. India is the latest country to report such activity.

Do anything? If you build software, check that no passwords or keys sit in public code or config files, and that cloud roles have only the access they need. These are old mistakes that AI agents find quickly.