Context

• Artificial Intelligence (AI), cyber capabilities, autonomous systems and data increasingly shape economic and military power.
• The convergence of AI and cyber threats is particularly dangerous because attacks can become faster, adaptive, scalable and autonomous.
• AI has evolved from an efficiency-enhancing tool into critical infrastructure for defence, intelligence, business and governance.
• AI-powered malware, autonomous agents and automated vulnerability discovery are challenging conventional cybersecurity systems and raising fundamental questions about human control and accountability.

A New and Disruptive Technological Era

• Large Language Models (LLMs) are expanding beyond information processing, while Agentic AI is increasingly capable of planning and executing tasks autonomously.
• Traditional cybersecurity relies heavily on predefined rules and predictable attack patterns. AI-powered malware can potentially analyse its environment, alter its behaviour and exploit emerging vulnerabilities.
• Advanced AI capable of identifying zero-day vulnerabilities can strengthen cyber defence but can equally be weaponised for offensive operations.
• The traditional Zero Trust security model may therefore require continuous adaptation as autonomous systems become capable of imitating legitimate behaviour and exploiting vulnerabilities faster than humans can respond.

AI and the Transformation of Warfare

• The Russia-Ukraine conflict and West Asian conflicts demonstrate AI’s growing military importance.
• AI-enabled systems can process satellite imagery, radio communications, electromagnetic signals and visual data to generate actionable intelligence.
• AI can improve target detection, missile tracking, battlefield intelligence, surveillance, reconnaissance and precision targeting.
• Autonomous weapons capable of identifying and engaging targets with limited human intervention could significantly alter military calculations.
• However, algorithms may recognise patterns without understanding their human, political and strategic context.
• Excessive dependence on automated assessments could therefore produce miscalculation and dangerous escalation.

The AI Power Shift

• AI is increasingly becoming a source of military, economic and technological power.
• States possessing superior computing capacity, data, semiconductors and advanced models could gain disproportionate strategic influence.
• The U.S.-China technological competition illustrates this emerging power struggle.
• The concentration of advanced AI capabilities within a few states and corporations creates risks of technological inequality and strategic dependence.
• At the same time, capabilities once restricted to nation states could become accessible to criminal networks or extremist organisations, potentially democratising offensive capabilities.

Cybersecurity in the Age of Agentic AI

• The transition from Generative AI to Agentic AI could dramatically transform cyber conflict. Autonomous systems may plan operations, interact with digital environments and adapt their behaviour.
• Attackers could use AI to discover vulnerabilities, generate malicious code and evade detection, while defenders deploy AI for continuous monitoring and automated response.
• This could create an AI-driven cyber arms race in which machine-to-machine confrontation occurs faster than humans can understand or control.

The Problem of AI Reliability

• AI’s growing capabilities do not eliminate hallucinations, algorithmic bias, erroneous reasoning and distorted outputs.
• These weaknesses become particularly dangerous in national security and military decision-making.
• An inaccurate AI-generated warning during a crisis could trigger escalation.
• AI can also contribute to algorithmic radicalisation, misinformation, deep fakes and political polarisation.
• A machine may identify a possible threat without possessing the contextual wisdom necessary to understand its significance.

Concentration of Technological Power

• The growing influence of major technology companies raises concerns over who controls frontier AI, determines acceptable uses and bears responsibility for harm.
• The U.S.-China AI rivalry demonstrates how technological development is increasingly connected with geopolitical competition.
• The race for supremacy could encourage states and corporations to prioritise speed and strategic advantage over safety.

The Need for Global AI Governance

• Human-in-the-Loop Decision-Making
  • AI should assist rather than independently determine decisions involving lethal force and strategic escalation. Meaningful human control must remain essential.
• International Norms
  • Major powers should establish common rules governing autonomous weapons, AI-enabled cyberattacks and strategic AI systems.
• AI Safety and Testing
  • Frontier models require rigorous testing for cyber vulnerabilities, manipulation, bias and unintended autonomous behaviour before deployment.
• Strengthening Cyber Resilience
  • Governments should modernise cybersecurity infrastructure, improve incident response and develop adaptive AI-based defence mechanisms.
• Transparency and Accountability
  • Clear responsibility must exist for decisions influenced by AI. Strategic technologies cannot operate without accountability and oversight.
• International Cooperation
  • AI governance must not become another arena of great-power rivalry. Cooperation among governments, technology companies, researchers and international organisations is essential.

India’s Strategic Imperative

• For India, AI-cyber convergence affects national security, critical infrastructure, defence modernisation and strategic autonomy.
• India should strengthen indigenous AI, semiconductor capabilities, cybersecurity research, defence applications and skilled manpower.
• India should promote an international framework based on responsible AI, human control, transparency and equitable access, while avoiding both technological complacency and excessive pessimism.

Conclusion

• AI can strengthen defence and resilience but can also enable autonomous attacks, cyber warfare, misinformation and destabilising military decisions.
• The objective should not be blind technological adoption or rejection.
• It should be cautious experimentation, strong safeguards, meaningful human control and continuous risk assessment.
• As AI increasingly determines the future of warfare and economic power, governance, wisdom and oversight must advance alongside technological innovation.