CERT-In warns of rising AI-driven cyber threats amid ‘Mythos’ concerns

New Delhi: India’s nodal cybersecurity agency, the Indian Computer Emergency Response Team (CERT-In), has issued a high-severity alert warning that rapid advances in frontier artificial intelligence are fundamentally reshaping the cyber threat landscape, raising fresh concerns across the technology ecosystem.

In an advisory titled “Defending Against Frontier AI Driven Cyber Risks”, CERT-In said the growing sophistication of next-generation AI systems is enabling attackers to operate with unprecedented speed, scale and autonomy.

The warning comes amid what industry insiders are calling “Mythos” jitters — a shorthand for anxiety over the disruptive and unpredictable capabilities of cutting-edge AI models.

According to the agency, modern AI tools can now independently identify vulnerabilities in widely used software, analyse massive volumes of source code, and execute complex, multi-stage cyberattacks with minimal human intervention.

This marks a sharp departure from traditional cyber threats, which typically required significant manual expertise and coordination.

CERT-In cautioned that such AI-driven systems can chain multiple exploits across different platforms, allowing attackers to infiltrate enterprise networks end-to-end.

By automating reconnaissance, exploitation and lateral movement, these tools significantly lower the barrier to entry for cybercriminals while amplifying the precision and impact of attacks.

The advisory highlighted that micro, small and medium enterprises (MSMEs) are particularly vulnerable due to limited cybersecurity infrastructure and resources.

It urged these organisations to strengthen their defences by deploying advanced threat detection systems, enabling continuous network monitoring, and maintaining detailed logs of system activity for effective forensic analysis.

Emphasising the growing speed at which vulnerabilities can be discovered and exploited, CERT-In called for robust patch management and timely disclosure practices.

Delays in addressing software weaknesses, it warned, could dramatically increase exposure to breaches in an AI-driven threat environment.

The agency also underscored the importance of secure coding standards and regular security audits, recommending that organisations adopt AI-aware security frameworks capable of anticipating adversarial use of machine learning technologies.

Log preservation, CERT-In noted, will be critical in responding to increasingly sophisticated attacks, as detailed records can help reconstruct intrusion pathways and limit damage.

IANS