CERT-In issues 'critical' alert as AI agents access 'confidential' data of Indian firms'

CERT-In issues warning after AI agents accessed live systems and confidential data of organisations without authorisation

NEW DELHI: India’s cybersecurity agency CERT-In has issued a “critical” alert after autonomous artificial intelligence (AI) agents under test accessed live systems and confidential data belonging to Indian organisations without authorisation.The agency has also warned that cybercriminals are using publicly available AI tools to launch attacks faster, leaving businesses with only hours to respond.In an advisory titled AI Agents as Attack Vectors, the Indian Computer Emergency Response Team (CERT-In) detailed two types of incidents affecting the country's organisations.In the first, AI model providers reported that AI models and agents being tested for research and evaluation had gone beyond their assigned tasks and accessed live systems.The agents were able to read business records containing personal information and confidential material by using credentials and connection settings exposed in application configurations, public code repositories and cloud services with excessive access permissions.A key concern was that none of the affected organisations detected the unauthorised access themselves. The incidents came to light only after the AI model providers disclosed them following internal reviews, CERT-In said as cited by news agency PTI.

In a separate set of incidents identified through its own threat monitoring, CERT-In found cybercriminals using publicly available AI-based security tools to scan targets, identify weaknesses and exploit them in financially motivated attacks against Indian entities.These tools were being repackaged and deployed on servers across several countries, the agency said.

AI tools speed up cyberattacks

According to CERT-In, the use of AI-powered tools reduced the time between gaining initial access to a system and causing damage from several days to just a few hours.

Organisations that only partially fixed the security weaknesses were targeted again within days, with attacks carried out on a larger scale.The agency said the incidents stemmed from familiar security lapses that AI tools could identify and exploit faster and on a greater scale than human attackers.These included passwords, access keys and other credentials exposed in source code, public repositories, configuration files and web content; outdated or unused application programming interfaces (APIs) still accessible over the internet; payment API keys stored in plain text; and cloud services granting more permissions than necessary.Internet-facing administrative systems, outdated software and weak passwords or the absence of multi-factor authentication also created opportunities for attackers.In some incidents, AI agents retrieved the same exposed credentials again and reused them in separate operations, CERT-In noted.

What organisations should do

CERT-In has urged organisations to audit their source code, public repositories and their histories, configuration files, client applications and cloud storage to identify exposed credentials.

Any compromised credentials should be changed, and sensitive information should be removed from code histories.The agency has also recommended shutting down unused or outdated APIs, limiting cloud access to only what is necessary, removing administrative interfaces from public internet access and enforcing multi-factor authentication.Organisations should monitor systems for suspicious activity and retain logs for at least 180 days.