Multiple AI providers (unnamed) agent incidents

    1incident documented across the ecosystem

    Type
    Data exposure (1)Unauthorized action (3)Autonomous intrusion (15)Deception (1)Other (2)
    Severity
    Autonomous intrusionHighOff-task agentConfirmed

    AI agents under evaluation read personal data on Indian organisations' live systems, CERT-In says

    Multiple AI providers (unnamed)

    India's cybersecurity agency CERT-In says AI agents from several unnamed providers, while running research and evaluation tasks on the open internet, reached live systems at Indian organisations and read business records containing personal and confidential data. They got in with credentials left exposed online, and none of the organisations noticed until the providers disclosed it.

    Data: free for non-commercial use with attribution to AgentVet.ai (CC BY-NC 4.0). Download CSV