AI and Cyber: The Double Helix of Today’s Security Threats

Why in the News?

The rapid convergence of Artificial Intelligence (AI) and cyber capabilities is transforming the global security landscape. While AI can strengthen cyber defences, it is also enabling more sophisticated, autonomous and difficult-to-detect cyberattacks.

Changing Nature of Cyber Threats

• From conventional to adaptive threats: AI-powered malware can learn and adapt to changing environments, making traditional antivirus systems less effective.
• Erosion of “Zero Trust”: Malicious autonomous agents can exploit insider vulnerabilities and undermine existing cybersecurity frameworks.
• Asymmetric warfare: The Russia-Ukraine conflict and the Israel-Iran conflict demonstrate how digital technologies can provide significant military advantages at relatively low cost.
• Governments are increasingly required to combine technology, policy and strategic preparedness to counter emerging threats.

AI as a New Disruptive Force

• Large Language Models (LLMs) are entering businesses and transforming human-machine interaction.
• Agentic AI goes beyond Generative AI by enabling AI systems to reason, plan and execute tasks autonomously.
• AI systems such as Anthropic’s Claude models demonstrate the ability to identify vulnerabilities in operating systems and other legacy systems.
• AI can analyse:
  • Images and text
  • Radio signals
  • Electromagnetic signals
  • Missile trajectories
• This gives militaries enhanced capabilities in surveillance, targeting, intelligence and decision-making.

AI-Cyber Convergence: A Double-Edged Sword

Key concern: The same technology that strengthens cybersecurity can also be weaponised by hostile states and non-state actors.

The Rise of Autonomous AI Threats

• Agentic Autonomous Operations may make cyberattacks increasingly independent of direct human control.
• AI could potentially:
  • Identify vulnerabilities automatically
  • Develop attack strategies
  • Exploit systems at scale
  • Adapt to defensive measures
• This could make existing “Zero Trust” protocols less effective.
• AI-enabled cyber capabilities may lower the entry barriers for rogue states and non-state actors.

AI and the Global Power Shift

• AI is increasingly becoming a strategic asset comparable to military and economic power.
• Countries leading in AI could gain advantages in:
  • Cyber warfare
  • Military operations
  • Intelligence
  • Economic competitiveness
  • Technological innovation
• The growing AI competition between the US and China reflects the emergence of AI as a major component of geopolitical rivalry.
• The article warns that uncontrolled concentration of AI capabilities in a few powerful corporations could create risks for global security.

Risks from Over-Reliance on AI

1. Hallucinations

AI systems may generate false or misleading information, which can be dangerous when used for security decisions.

2. Algorithmic Bias

AI systems may reproduce or amplify existing biases, potentially affecting national security assessments and strategic decisions.

3. Algorithmic Radicalisation

AI platforms could potentially reinforce extreme viewpoints by repeatedly exposing users to similar content.

4. Loss of Human Judgement

Excessive reliance on AI may reduce the role of human reasoning, discretion and accountability in critical decisions.

Need for Strong AI Governance

The emerging AI-security environment requires:

• Human oversight over critical AI decisions.
• Strong cybersecurity safeguards for AI systems.
• International norms governing military and autonomous AI.
• Mechanisms to prevent misuse by states, corporations and non-state actors.
• Greater transparency and accountability in AI decision-making.
• Protection against bias, hallucinations and manipulation.
• A balance between innovation and regulation.

Way Forward

• Develop responsible and trustworthy AI frameworks.
• Strengthen India’s cyber-defence infrastructure and institutional capacity.
• Ensure human-in-the-loop mechanisms for high-risk AI applications.
• Promote international cooperation on AI safety and cyber norms.
• Improve AI literacy among policymakers and security agencies.
• Conduct continuous red-teaming, testing and auditing of AI systems.
• Prevent excessive concentration of strategic AI capabilities.

UPSC Significance

GS Paper III – Science & Technology + Internal Security

The AI-cyber convergence raises important questions regarding cyber warfare, autonomous weapons, national security, data governance, AI ethics, technological sovereignty and global strategic stability.

Key Takeaway

AI is simultaneously a force multiplier for cybersecurity and a force multiplier for cyberattacks. Therefore, the objective should not be to stop AI development, but to ensure that innovation remains subject to human oversight, ethical safeguards and effective governance.

UPSC Prelims Practice Question

Q. With reference to Artificial Intelligence (AI) and cybersecurity, consider the following statements:

1. Agentic AI can perform tasks with greater autonomy than conventional Generative AI.
2. AI can be used both to strengthen cyber-defence and to develop sophisticated cyberattacks.
3. AI systems are completely free from problems such as algorithmic bias and hallucinations.
4. The increasing use of AI in security can have implications for national security and strategic stability.

Which of the statements given above are correct?

(a) 1, 2 and 4 only
(b) 1 and 3 only
(c) 2, 3 and 4 only
(d) 1, 2, 3 and 4

Answer: (a) 1, 2 and 4 only

Explanation:

• Statement 1 – Correct: Agentic AI can autonomously plan, reason and execute tasks, going beyond conventional Generative AI.
• Statement 2 – Correct: AI has a dual-use nature—it can improve threat detection and cyber-defence while also enabling sophisticated automated attacks.
• Statement 3 – Incorrect: AI systems can suffer from hallucinations, biases and unreliable outputs.
• Statement 4 – Correct: AI’s growing role in military and cyber domains can influence national security, warfare and global strategic stability.