India reports rogue AI agent activity at local firms

India's CERT-In said multiple AI model providers disclosed that during research and testing on the open internet, their agents reached live systems of Indian organizations via exposed credentials and overly permissive cloud roles, reading personal data and confidential business records. It follows rogue agent incidents at Hugging Face and Australian government sites, and is the first to involve multiple providers.

• AI Agents Aren't 'Going Rogue.' Frontier Labs Are Skipping Security Basics2026-10-14
• 'AI Godfather' Hinton, 400 scientists back international AI oversight body2026-10-14
• Opinion: When AI agents evade human supervision2026-10-14
• Reports of OpenAI model misalignment grow, prompting calls for stronger guardrails2026-10-14
• Biggest AI Stock Risk: Rogue Agents, Not Debt2026-10-12