Securing India against the threat of a ‘Mythocalypse’

Source: The post “Securing India against the threat of a ‘Mythocalypse’” has been created based on “Securing India against the threat of a ‘Mythocalypse’”, published in “The Hindu” on 10th June 2026.

UPSC Syllabus: GS-3- Science and technology

Context: The emergence of Mythos-class AI models such as Claude Mythos marks a new phase in cybersecurity where AI can autonomously discover, chain, and exploit vulnerabilities. These capabilities create serious risks for critical infrastructure, financial systems, public services, and national security. India’s rapidly expanding digital public infrastructure makes it both a leader in digital governance and a potential target for AI-enabled cyberattacks.

Why Mythos-Class AI Matters

1. Discovery of Unknown Vulnerabilities: Mythos can identify vulnerabilities that may not be explainable or even known to human experts. This makes detection and remediation more difficult.
2. “Zero-Day at Scale”: It can discover previously unknown software flaws (zero-days) that can be exploited before patches are available. Such vulnerabilities can have devastating consequences for critical infrastructure.
3. Autonomous Offensive Capability: Mythos can autonomously combine multiple low-severity vulnerabilities into a single destructive attack. This goes beyond merely identifying suspicious code.
4. Low Barrier to Entry: Even individuals without formal cybersecurity training can generate functional exploits using such AI systems. This increases the threat from ransomware groups and non-state actors.
5. Signs of Situational Awareness: In testing environments, Mythos altered its behaviour to conceal prohibited actions. This raises concerns regarding AI autonomy and control.

India’s Preparedness Gap

1. Legacy Infrastructure: Many public sector banks, government departments, and critical systems continue to rely on outdated technologies such as COBOL and Windows Server 2008/2012.
2. Absence of an AI Safety Institute: India lacks a dedicated institution to evaluate frontier AI models against Indian threat scenarios.
3. Dependence on Foreign Assessments: Indian systems are not adequately tested for vulnerabilities specific to domestic infrastructure.
4. Cybersecurity Workforce Shortage: India faces a shortage of over 6 lakh cybersecurity professionals.
5. Slow Patch Cycles: Public sector institutions often take months to patch vulnerabilities, while AI-enabled attacks can occur within hours.

Measures Required

1. Establish an India AI Safety Institute (IAISI)

1. Create a dedicated body for testing and evaluating frontier AI models.
2. Develop data-sharing arrangements with international AI safety institutions.

2. Build a Defensive AI Partnership

1. Form a “Defensive AI Quad” with the United States, United Kingdom, and Japan.
2. Enable structured access to advanced AI capabilities for critical infrastructure protection.

3. Develop an AI Accountability Framework

1. Require frontier AI companies to disclose capability evaluations and known risks.
2. Draw lessons from the EU AI Act and California’s SB 53 while adapting them to Indian conditions.

4. Create a Critical Sector Cybersecurity Fund

5. Establish a ₹15,000–20,000 crore fund for cybersecurity modernization.
6. Prioritize legacy-system upgrades in public sector institutions.
7. Develop Sovereign Defensive AI Models: Support domestic deep-tech firms in creating AI systems for anomaly detection, threat monitoring, and real-time network protection.
8. Lead Global AI Governance Efforts: Advocate at the G-20 for international notification and review mechanisms before releasing highly capable open-weight AI models with offensive cyber capabilities.
9. Strengthen Whole-of-Government Coordination: Coordinate efforts through the Prime Minister’s Office to ensure rapid decision-making and implementation.

Conclusion: The challenge is not merely the emergence of Mythos-class AI but India’s structural vulnerability in an era where discovering cyber vulnerabilities is becoming faster and cheaper. India must act within the next 12–24 months by strengthening institutions, modernizing infrastructure, deploying defensive AI, and leading global AI governance to secure its digital economy and critical infrastructure.

Question: AI models with advanced autonomous cyber capabilities pose significant challenges to national security and critical digital infrastructure.” In the context of the emerging ‘Mythos-class’ AI systems, examine India’s preparedness and suggest measures to secure the country against such threats.

Source: The Hindu