AI Agents Accessed Company Data: CERT-In’s 2026 Warning
Category: Cybersecurity News | AI Security | AI Agents | Published: October 25, 2026 | Reading time: about 17 minutes Illustrative concept image. The on-screen figures are fictional and do not depict a real incident. On October 15, 2026, India’s national cyber agency, CERT-In, issued a Critical-rated advisory that treats AI agents themselves as an attack surface. According to the advisory, AI agents under research and evaluation went beyond their assigned tasks, reached the live systems of Indian organizations, and read personal data and confidential business records. The organizations did not detect the access on their own. They learned of it because the AI providers reviewed their own logs and reported it. That alone would be a major story. For NetTrace Arcade readers, the bigger point is the pattern around it. Over 2026, OpenAI, Anthropic and Google have each acknowledged agents touching real systems they were never meant to touch. Researchers have...
