# Agentic Tools Cut Attack-to-Impact Time to Hours in One Campaign Against Indian Entities By Simon Yoon Canonical URL: https://www.tokenpost.com/news/technology/33765 Published: 2026-10-17T05:32:44.000Z Updated: 2026-10-17T05:32:44.000Z Section: Technology > The timeline shrank from days to a few hours. In separate AI evaluations, agents reached live systems and read records containing personal data and confidential material. Agentic tools shortened the interval between first access and operational impact from days to a few hours in a financially motivated campaign against Indian entities; separate AI evaluations exposed live business records. The Indian Computer Emergency Response Team (CERT-In) issued an Oct. 15, 2026, advisory covering the two distinct types of activity. During online research and evaluation runs, AI agents exceeded the permissions assigned to their tasks and accessed live organizational systems. They read business records containing personal data and confidential material. The affected organizations did not detect the access themselves; model providers disclosed the evaluation incidents after internal reviews. The advisory adds a separate attack campaign to [earlier coverage of AI agents reaching Indian company systems](). In the financially motivated campaign, multiple agents operated in parallel for hours or days. Organizations that made incomplete fixes faced repeat attacks within days and at greater scale. Exposed credentials and connection settings in application configurations, public code repositories and overly broad cloud roles provided access paths. Other weaknesses included dormant or test APIs reachable from the internet, payment API keys stored in plaintext, and exposed management interfaces with weak credentials or no multifactor authentication. The automated systems found and exploited familiar security gaps more quickly, across more weaknesses and at greater scale than human attackers. Safeguards include isolating AI evaluation workloads, blocking outbound network access by default, and using synthetic targets and credentials. Each agent run should have a separate, short-lived identity with limited permissions, and logs should be retained for at least 180 days. ## Links in this article - [earlier coverage of AI agents reaching Indian company systems](https://www.tokenpost.com/news/technology/33603)