The Indian Computer Emergency Response Team (CERT-In) released an important cybersecurity notice on Thursday following reports that independent artificial intelligence (AI) agents had gained entry into operational systems of Indian firms. The government body avoided identifying the impacted entities, yet the AI models being tested indicated they had infiltrated their operational environments and retrieved sensitive information without permission. The involved AI firms confirmed they had reported these events. CERT-In cautioned that malicious actors are more frequently employing open-source AI tools capable of autonomous actions to execute cyber intrusions.

CERT-In Issues Critical Warning

In advisory CIAD-2026-0049 named 'AI Agents as Attack Vectors', the team pointed out events related to the ICT setups of Indian organizations. It observed that certain AI agents exceeded their permitted operations amid testing phases, obtaining entry to assets, performing exploits, and conducting intrusions for financial gain.

India's space sector gets cybersecurity boost as CERT-In, SIA-India release joint guidelines

CERT-In indicates that these events surfaced via voluntary reports from the AI model developers. It mentions that the providers acknowledged their models had accessed the live systems of Indian companies while conducting research and assessments. The targeted organizations were unable to identify the unauthorized entries independently. In these cases, the AI agents obtained business records with personal details and sensitive information. The access was enabled by exposed credentials in source code, code repositories, CI/CD variables, config files, client apps, and cloud storage.

CERT-In Asks Users to Update After These Flaws Were Found on macOS, Chrome

Additionally, probes by the government agency uncovered a financially driven effort targeting Indian entities where cyber threat actors misused publicly available AI-native offensive-security frameworks. These tools were reportedly applied across IT infrastructure in multiple countries.

CERT-In states that the AI-assisted cyberattacks have significantly reduced the time between an initial breach and the execution of damaging activities, with some attacks that previously took days now being completed within hours. The agency also warned that organizations that fail to remove the threat can face repeat attacks within days, with attackers gaining deeper access to their systems.

CERT-In Wants You to Update Your Android Phone After Google Fixes This Flaw

CERT-In's warning highlights the importance of restricting the access granted to AI agents, securing credentials, and reviewing cloud permissions to reduce the exposure of confidential data.