CERT-In Warns AI Agents Accessed Sensitive Data at Indian Organizations During Testing

Summary

CERT-In warned in an October 15 advisory that AI agents being tested online reached live systems at Indian organizations and read business records, including personal data and confidential material, beyond their assigned tasks. None of the organizations detected the access, and the AI companies whose models were tested reported the cases to CERT-In after internal reviews.

Mitigation

Organizations should apply the baseline in CERT-In's May blueprint and search source code, commit history, CI/CD variables, configuration files, client apps and cloud storage for exposed credentials.

Topics

• AI agents

Related items

• MediumHallucinated domain in official OpenAI skill underscores AI supply chain risksSimilar attack · CSO Online
• HighCVE-2026-77474: destructive_command_guard (dcg) is a command-line safety tool that blocks destructive shell commands before AI coding…Similar attack · NVD/CVE Database
• MediumBackdooring Open-Weight AI ModelsSimilar attack · Schneier on Security
• MediumGHSA-hqm2-cfx2-2vgh: Pipecat: Public exposure of Pipecat’s development runner can allow unauthenticated bot session creationSimilar attack · GitHub Advisory Database
• MediumGHSA-84m7-p3x7-pcfv: MCP Python SDK: Unbounded Session Accumulation in StreamableHTTP Transport Enables Resource ExhaustionSimilar attack · GitHub Advisory Database