How is militarization of Artificial Intelligence and Computing revolutionizing the military affairs?

Introduction

Speaking to Russian students on 1 September 2017, President Vladimir Putin said that whoever becomes the leader in artificial intelligence "will become the ruler of the world." Military planners in Washington, Beijing and elsewhere share this belief. Artificial intelligence (AI), together with advances in computing power, data and sensors, is changing how wars are planned, fought and won, much as gunpowder, the aircraft and nuclear weapons did before. From Ukraine's drone war to AI-assisted targeting in Gaza, this transformation is already visible, and it raises profound questions of strategy, ethics and stability.

Overview

The idea of a Revolution in Military Affairs (RMA) gained prominence after the 1991 Gulf War, when precision-guided weapons and information networks gave the US a decisive edge. Today a new revolution is under way. The US Department of Defense launched Project Maven in 2017 to use AI to analyse drone video, and in 2023 announced the Replicator initiative to field thousands of cheap autonomous systems. China's 2019 defence white paper described a shift towards "intelligentised warfare". Russia, Israel, Turkey, Iran and India are investing heavily. AI's military uses range from data analysis to fully autonomous weapons.

How AI and computing are revolutionising military affairs

1. Intelligence, surveillance and reconnaissance.

AI can scan satellite images, drone feeds and intercepted communications far faster than human analysts, detecting targets and patterns in real time. (Reference: US Department of Defense, Project Maven memorandum, 26 April 2017; RAND Corporation, The Department of Defense Posture for Artificial Intelligence, 2019.)

2. Faster decision-making.

AI-enabled command systems fuse data from all domains to give commanders recommendations in seconds, compressing the "observe-orient-decide-act" cycle, as in the US concept of Joint All-Domain Command and Control (JADC2). (Reference: US Department of Defense, Summary of the JADC2 Strategy, March 2022.)

3. Autonomous weapons and drone swarms.

Loitering munitions and swarms of drones can find and attack targets with little human input. Ukraine and Russia use millions of drones, including AI-assisted ones, reshaping the battlefield. (Reference: Royal United Services Institute (RUSI), reports on drone warfare in Ukraine, 2023–2024; US Deputy Secretary of Defense, Replicator announcement, August 2023.)

4. AI-assisted targeting.

Militaries use AI systems to generate targets at a speed humans cannot match; investigative reports on systems such as "Lavender" and "The Gospel" in Gaza raised alarm about civilian harm. (Reference: +972 Magazine and Local Call investigation, April 2024; statements of the UN Secretary-General on AI in targeting, April 2024.)

5. Cyber and electronic warfare.

AI can find vulnerabilities, launch automated cyber attacks and defend networks, while electronic warfare systems use machine learning to jam and spoof enemy signals. (Reference: NATO Cooperative Cyber Defence Centre of Excellence publications; the Stuxnet case, 2010.)

6. Logistics and predictive maintenance.

AI predicts equipment failures, manages supply chains and optimises fuel and ammunition, increasing readiness and reducing costs. (Reference: US Air Force, predictive maintenance programmes; NATO, Artificial Intelligence Strategy, 2021.)

7. Training and simulation.

AI-powered simulators create realistic adversaries; in DARPA's AlphaDogfight Trials (2020), an AI agent defeated an experienced F-16 pilot 5–0 in simulated combat. (Reference: Defense Advanced Research Projects Agency (DARPA), AlphaDogfight Trials, August 2020.)

8. Quantum and space computing.

Quantum computing could break current encryption and improve sensing, while AI manages satellite constellations for communication and targeting, as shown by the use of commercial satellites in Ukraine. (Reference: NATO, Quantum Technologies Strategy, 2024; CSIS, Space Threat Assessment 2024.)

Risks and challenges

1. Loss of human control.

Fully autonomous weapons could select and kill targets without meaningful human judgment, raising moral and legal concerns. (Reference: International Committee of the Red Cross (ICRC), Position on Autonomous Weapon Systems, May 2021.)

2. Escalation and nuclear risks.

Machine-speed decisions could cause accidental escalation, and AI in early-warning or nuclear command systems could produce catastrophic errors. In November 2024, the US and Chinese presidents agreed that humans, not AI, should control decisions to use nuclear weapons. (Reference: White House readout of the Biden-Xi meeting, Lima, 16 November 2024; SIPRI, Artificial Intelligence, Strategic Stability and Nuclear Risk, 2020.)

3. Accountability gaps.

If an autonomous system commits a war crime, it is unclear whether the commander, programmer or manufacturer is responsible. (Reference: Human Rights Watch and Harvard Law School, Mind the Gap: The Lack of Accountability for Killer Robots, 2015.)

4. Proliferation to non-state actors.

Cheap drones and open-source AI allow terrorist groups to conduct precise attacks, as groups such as the Houthis have shown with drones and missiles. (Reference: UN Panel of Experts on Yemen reports; UN Counter-Terrorism Committee Executive Directorate, Delhi Declaration, 2022.)

5. Bias and errors.

AI systems trained on flawed data can misidentify civilians as combatants and are vulnerable to hacking and spoofing. (Reference: UNIDIR, Algorithmic Bias and the Weaponization of Increasingly Autonomous Technologies, 2018.)

6. An AI arms race.

Competition between the US and China in AI and chips, including US export controls of October 2022, fuels a race in which safety may be sacrificed for speed. (Reference: US Bureau of Industry and Security export control rules, 7 October 2022; National Security Commission on Artificial Intelligence, Final Report, 2021.)

7. Erosion of international humanitarian law.

Principles of distinction, proportionality and precaution are harder to apply when decisions are made by algorithms at high speed. (Reference: Geneva Conventions, Additional Protocol I (1977), Articles 48, 51 and 57.)

8. A widening gap between militaries.

Countries with advanced AI and computing gain overwhelming advantages, leaving developing states more vulnerable and dependent on technology imports. (Reference: Stanford University, AI Index Report 2024; SIPRI, military expenditure data.)

Governance efforts and Pakistan's position

Governments have discussed lethal autonomous weapons systems (LAWS) under the Convention on Certain Conventional Weapons since 2014, through a Group of Governmental Experts since 2017. The UN General Assembly adopted its first resolution on autonomous weapons in December 2023, and the US launched a Political Declaration on Responsible Military Use of AI and Autonomy in 2023. Pakistan was among the first states to call for a ban on fully autonomous weapons at the CCW in 2013, arguing that they lower the threshold for war. At the same time, Pakistan must invest in AI for defence, cyber security and surveillance, as its National AI Policy (2025) and the use of drones and electronic warfare in the May 2025 conflict with India showed.

Conclusion

Stephen Hawking warned in 2014 that "the development of full artificial intelligence could spell the end of the human race." In military affairs, AI and computing are already revolutionising intelligence, decision-making, weapons, cyber warfare and logistics, giving advantages to those who master them. But they also bring the risk of machines making life-and-death decisions, uncontrolled escalation and new arms races. The challenge for the world, and for Pakistan, is to harness AI for security while ensuring meaningful human control and rules that keep war within the limits of law and humanity.