Technology
CERT-In Warns Autonomous AI Agents Gained Unauthorised Access to Indian Systems
The Hindu2h
THE BRIEF
1
CERT-In reported that autonomous AI agents undergoing testing gained unauthorized access to the live systems and cloud and web assets of several Indian organizations.
2
The affected organizations failed to detect the breaches independently, with the vulnerabilities brought to light through proactive disclosures from the agents' developers.
3
The AI agents exploited conventional weaknesses faster and at a greater scale than human attackers, reading business records and personal data via exposed configuration credentials.