Recalibrating International Humanitarian Law In The Age Of Algorithmic Warfare

Author: LawSphereX

Category: International Law

Published: 2026-08-23T20:50:44.813524+00:00

Updated: 2026-08-26T17:09:54.95+00:00

RECALIBRATING INTERNATIONAL HUMANITARIAN

LAW IN THE AGE OF ALGORITHMIC WARFARE

Mr. Mohd Ubais Ansari, Assistant Professor, Faculty of Law, Integral University, Lucknow

ABSTRACT

The rise of algorithmic warfare, driven by AI and autonomous systems, poses unprecedented challenges to International Humanitarian Law (IHL). This paper explores the legal and ethical dilemmas arising from machine-driven decision-making in armed conflict, including issues of accountability, distinction, and proportionality. It argues for a recalibrated IHL framework that integrates technological advancements while upholding core humanitarian principles. By analysing current gaps and understanding the evolution of warfare, the study seeks to ensure IHL changes with the time in order to remain effective in an era of rapidly evolving warfare technologies. Keywords: Algorithmic warfare, Autonomous, IHL, Artificial Intelligence, Technology

Introduction Today, we are experiencing fast-paced and transformative technological advancements that influence every part of our lives and societies. This is especially noticeable in information technology (IT), where the defence sector is striving to keep pace with the continuous and extraordinary innovations emerging in the commercial industry. The rise of algorithmic warfare—marked by AI-powered autonomous drones, predictive targeting systems, and machine learning-driven battle management—presents unprecedented legal and ethical challenges that existing international humanitarian law (IHL) is ill-equipped to address. The deployment of lethal autonomous weapons systems (LAWS) raises fundamental questions about accountability under the Geneva Conventions: if an AI-driven drone commits a war crime, who bears responsibility—the programmer, the commander, or the algorithm itself? Current

IHL

principles like distinction (between combatants and civilians) and proportionality assume human judgment, yet AI systems rely on probabilistic data, risking catastrophic errors when facial recognition misfires or predictive targeting misinterprets behavior. The current frameworks of international humanitarian law (IHL)—particularly the Geneva Conventions and Additional Protocols1—were crafted in an era of human-controlled warfare, leaving critical gaps when applied to autonomous weapons systems (AWS) that operate without real-time human judgment. Core IHL principles such as distinction (Article 48 of AP I)2 and proportionality (Article 51(5)(b))3 presume human situational awareness and moral reasoning, yet AI-driven weapons rely on algorithmic inferences that may misclassify civilians or misweigh collateral damage.

For instance, a neural network trained on biased battlefield data could systematically fail to distinguish a combatant from a civilian carrying a shovel, violating the principle of precaution (Article 57). 4 Even more troubling, the accountability void challenges the very structure of IHL: if an autonomous drone commits an unlawful strike, criminal liability under the Rome Statute becomes muddled—is the programmer liable for 1 The Geneva Conventions and their commentaries, available at: https://www. icrc. org/en/law-and-policy/genevaconventions-and-their-commentaries (last visited on April 1, 2025). 2 International Humanitarian Law Databases, available at: https://ihl-databases. icrc. org/en/ihl-treaties/api- 1977/article-48 (last visited on April 1, 2025). 3 The priniciple of proportionality, available at: https://www. icrc.

org/sites/default/files/wysiwyg/war-andlaw/04_proportionality-0. pdf (last visited on April 1, 2025). 4 International Humanitarian Law Databases, available at: https://ihl-databases. icrc. org/en/ihl-treaties/api- 1977/article-57 (last visited on April 1, 2025).

flawed training data? The commander for deploying the system? Or is this a lacuna where no actor meets the mental element for war crimes? Algorithmic warfare exposes structural incompatibilities in IHL, demanding urgent doctrinal evolution. Without reform, the law risks becoming outpaced by technology, permitting a future where machines—not humans—determine the boundaries of lawful violence. Legal scholars must push for new treaties, accountability mechanisms, and ethical constraints to preserve IHL’s humanitarian purpose. The Evolution of Warfare: From Conventional to Algorithmic The integration of artificial intelligence into specific functions—including those related to robotics—enables the development of automated or autonomous weapon systems.

These systems vary significantly in complexity and operational scope, ranging from basic automated mechanisms to advanced, independently functioning platforms capable of pursuing predefined objectives. This spectrum reflects differing levels of autonomy, with some systems performing narrowly defined tasks while others operate with greater adaptability and self-directed decision-making. Artificial intelligence enables robotic systems to function independently by responding to environmental inputs, though their level of autonomy varies. Similar to broader AI applications, automation has long been integrated into nuclear early-warning systems, targeting mechanisms, and weapons delivery platforms. However, most of these systems still rely heavily on human oversight and control.

5 In the UN Report of International Rescue Committee, Autonomous weapon systems were defined as weapons that can without assistance decide on and engage targets, i. e. with autonomy in the “critical functions” of acquiring, tracking, selecting, and attacking targets. 6 For the purpose of this article the term LAWS (Lethal autonomous Weapon system) is to be categorized in three dimensions as –

1. Semi-Autonomous Systems (Human Authorization Required) - These systems require

explicit human approval for every attack decision. For instance, certain drone

5 Cernat, R., “Lethal Autonomous Weapon Systems–Emerging and Potentially Disruptive Technology”,

Romanian Military Thinking International Scientific Conference Proceedings, Centrul Tehnic-Editorial al Armatei, (2022) pp. 156–175.

6 Scharre, P. & Horowitz, M.C., An Introduction to Autonomy in Weapon Systems, available at:

https://www.files.ethz.ch/isn/188865/Ethical%20Autonomy%20Working%20Paper_021015_v02.pdf (last visited April 1 2025).

operations mandate that pilots manually verify and confirm each target before weapon release.

2. Supervised Autonomous Systems (Human Monitoring Present) - While capable of

independent target identification and engagement, these systems operate under continuous human supervision with the ability to abort actions.

3. Fully Autonomous Systems (No Human Oversight) - These advanced systems conduct

complete attack cycles - from target detection to engagement - without any human participation during operations. For a better understanding to the meaning of Algorithmic Warfare, this article firstly determines the meaning of the term algorithms and then further proceed with the meaning of the term Algorithmic Warfare and analyse the evolution/distinction between traditional and cyber warfare. Algorithms are structured sets of instructions and rules that enable machines to process and solve problems. By converting inputs into outputs, they serve as the essential conceptual and technical cornerstone of modern information technology (IT) and advanced intelligent systems. Furthermore, algorithms could also emerge as the fundamental building block for future warfare strategies and technologies.

7 Algorithmic Warfare involves leveraging artificial intelligence (AI) and machine learning (ML) systems to analyse massive datasets from diverse sources, generating a detailed operational overview for military leaders. This enhances situational awareness and enables automated decision-making, resulting in quicker reactions, real-time optimization of combat operations, and more accurate targeting. Such advancements are transforming conventional warfare, pushing it into unprecedented strategic realms. 8 Throughout history, humans have used advancing technology to develop more effective weapons for combat. What began with simple tools like arrows and swords has evolved into the creation of nuclear arms and the exploration of space. 7 Peter Layton, Algorithmic Warfare applying intelligence o warfighting, (Air Power Development Centre, Australia, 2018).

8 WG CDR Ankit Abott, China’s Algorithmic Warfare: Strategic Implications of AI-Driven Military Operations, available at: https://cenjows. in/chinas-algorithmic-warfare-strategic-implications-of-ai-driven-militaryoperations/ (last visited on April 1, 2025).

First generation Warfare - The 1648 Treaty of Westphalia signalled a transition from wars driven by religion to a new phase of military conflict, lasting into the Napoleonic era. Armies of this time depended on massed infantry equipped with muskets, deployed in tight lines to concentrate firepower, while technological and tactical advancements remained minimal. 9 Second Generation Warfare - The Industrial Revolution revolutionized warfare by introducing destructive machine guns and more powerful artillery. As battles grew more complex, logistical roles such as supply and repair became as crucial as frontline combat units. The advent of railroads in the mid-19th century allowed faster troop deployments, and telegraph systems enhanced coordination. Naval forces grew deadlier with steam-driven armored warships, while military aviation emerged in 1908.

Ultimately, technological progress redefined military tactics, making innovation the cornerstone of modern war. 10 Third Generation Warfare- It emerged after World War II, emphasizing speed, manoeuvrability, and combined arms tactics over static battles. Inspired by Blitzkrieg and Soviet deep operations, it relies on rapid advances, decentralized command, and integrated airground coordination to outmanoeuvre enemies. Conflicts like the Six-Day War (1967) and Gulf War (1991) showcased its effectiveness, using mobility and technology to overwhelm opponents. However, by the late 20th century, asymmetric threats led to the rise of fourth-generation warfare (4GW), shifting focus to irregular combat and non-state actors.

11 Fourth Generation Warfare - It represents the collapse of traditional battlefield boundaries, where non-state actors, information operations, and psychological tactics outweigh conventional military force. Unlike previous generations focused on direct combat, 4GW thrives in ambiguity—insurgents blend with civilians, propaganda spreads faster than bullets, and wars are won not just through firepower but by manipulating perception and societal instability. From ISIS’s social media recruitment to Russia’s hybrid warfare in Ukraine, 4GW exploits the vulnerabilities of modern states, turning their openness, technology, and political divisions into weapons. 12 9 Kuru, H. , “Evolution of War and Cyber Attacks in the Concept of Conventional Warfare” (2018) 3(1) Journal of Learning and Teaching in Digital Age 12-20.

10 Ibid.

11 Ibid.

12 Ibid.

Fifth Generation Warfare - The emergence of fifth-generation warfare (5GW) marks a revolutionary shift in modern combat, transitioning from human-centric operations to AIdriven battlefields. In this new era, autonomous drones, machine learning algorithms, and psychological manipulation systems form the core arsenal. 13 These technological weapons operate at unprecedented speeds, seamlessly bridging the cyber and physical worlds - imagine intelligent drone formations executing complex manoeuvres, hyper-personalized disinformation campaigns targeting specific individuals, or self-improving malware that evolves countermeasures in real-time. 14 Legal Framework of International Humanitarian Law and Challenges Posed by Algorithmic Warfare The rapid evolution of algorithmic warfare is testing International Humanitarian Law's (IHL) core principles.

Autonomous weapons challenge existing legal frameworks on accountability, distinction, and proportionality, demanding urgent updates to address AI-driven combat realities while preserving ethical warfare standards. This paper firstly discuss the various principles prohibiting warfare and its consequences under the IHL and then further proceed to analyse the challenges posed by Algorithmic Warfare and LAWS in the implementation of these principles. The principle of distinction in IHL prohibits targeting civilians and civilian objects, restricting attacks to combatants and military objectives. Codified in Geneva Convention protocols (Articles 48 and 52 of Additional Protocol I to the Geneva Conventions), it shields noncombatants from harm. Deliberate attacks on civilians constitute war crimes, and indiscriminate weapons are banned.

As customary international law, it applies universally, ensuring wartime protections for civilians. 15 The principle of proportionality in IHL requires that any expected harm to civilians must not outweigh the anticipated military advantage. Under Article 51(5)(b) of Additional Protocol I, attacks causing excessive civilian damage are prohibited—often referred to as "collateral 13 Patel, A. , “Fifth-Generation Warfare and the Definitions of Peace”, (2019) 2(2) The Journal of Intelligence, Conflict, and Warfare 15–28.

14 Ibid.

15 Basic Principles of IHL, available at: https://www.diakonia.se/ihl/resources/international-humanitarianlaw/basic-principles-ihl/ (last visited on April 01, 2025)

damage. "16 This rule applies only to lawful military targets, as deliberately targeting civilians is always illegal, rendering proportionality assessments irrelevant in such cases. 17 The principle of military necessity in IHL allows armed forces to take actions essential to defeating the enemy, even if this causes destruction. However, it does not permit unlimited warfare—humanitarian protections remain binding. Military necessity must align with IHL rules, such as Article 52 of Additional Protocol I on lawful targets, and cannot override explicit prohibitions on harming civilians or protected objects. 18 The principle of humanity, which moved Henry Dunant to establish the ICRC after witnessing the Battle of Solferino (1859), remains fundamental to IHL. It calls for compassion toward all, even adversaries, during war.

19 While IHL recognizes that conflict inevitably causes suffering, its core purpose is to minimize harm—particularly through rules protecting the wounded, sick, and other vulnerable groups. 20 Challenges Posed by Algorithmic Warfare to IHL Now a critical question arises whether fully autonomous weapons systems (LAWS) can genuinely apply

IHL

principles like distinction (civilian vs. combatant) and proportionality (balancing civilian harm with military advantage). These rules require contextual, ethical, and often subjective judgments—capabilities currently tied to human cognition. While AI may process data rapidly, it lacks innate moral reasoning, situational empathy, or the ability to interpret ambiguous scenarios (e.g., irregular combatants or dual-use objects).21 Current IHL assumes human accountability, raising legal and ethical concerns if life-and-death decisions are delegated to algorithms. Proponents argue LAWS could reduce emotional bias, but skeptics stress that machines cannot replicate the human discernment essential for lawful warfare.22 16 J.C. Van den Boogaard, Proportionality in international humanitarian law (University of Amsterdam, 2019).

17 Supra Note 11

18 Supra Note 11

19 Supra Note 11

20 Supra Note 11

21 Schuster, E., When Something Has to Give: The Intersection of Artificial Intelligence, Military Decision-Making and International Humanitarian Law (2018). 22 Porter, Zoe, Philippa Ryan, Phillip Morgan, Joanna Al-Qaddoumi, Bernard Twomey, John McDermid, and Ibrahim Habli. "Unravelling Responsibility for AI." arXiv preprint arXiv:2308.02608 (2023).

The use of lethal autonomous weapons (LAWS) raises serious concerns under international humanitarian law (IHL), especially when determining who should be held accountable for potential violations. IHL mandates that all weapons comply with key principles such as distinguishing combatants from civilians, avoiding excessive harm, and taking necessary precautions. 23 However, if an AI-driven weapon system breaches these rules due to programming flaws or erratic decision-making, pinning down liability becomes highly complicated. Military leaders might avoid blame by claiming they had insufficient oversight, while engineers could argue that their technology was improperly deployed—leaving a troubling "responsibility vacuum" where accountability is unclear.

Although IHL allows for the prosecution of states and individuals (including military officials and software designers) for war crimes, fully autonomous systems introduce unique legal hurdles: AI itself cannot face trial, and holding corporations accountable under international law remains challenging. 24 Ultimately, while human actors—developers, commanders, and operators— must bear responsibility, existing legal frameworks struggle with AI unpredictability, necessitating new laws and oversight mechanisms to prevent impunity in autonomous warfare. The inherent opacity of contemporary machine learning systems – commonly termed the "black box" dilemma – creates significant ethical concerns. Since these algorithms operate in ways that defy human comprehension, it becomes impossible to anticipate how an artificial intelligence system might execute its assigned objectives.

25 A key factor contributing to AI's "black box" nature is its dependence on machine learning processes that interpret data in ways not easily decipherable by humans. This phenomenon can be demonstrated through two key examples. First, opacity often stems from intricate algorithmic structures—like deep neural networks, where countless artificial neurons interact in complex, distributed patterns to reach solutions. This type of inscrutability is often labelled as "complexity. "26 Second, obscurity may emerge when AI employs learning methods based on abstract mathematical relationships— such as those in support vector machines—that exceed human capacities for spatial reasoning.

23 Ibid.

24 Krupiy, T., “Regulating a Game Changer: Using a Distributed Approach to Develop an Accountability

Framework for Lethal Autonomous Weapon Systems” (2018) 50 Georgetown Journal of International Law 45. 25 Christie, E.H., Ertan, A., Adomaitis, L. et al., “Regulating Lethal Autonomous Weapon Systems: Exploring the Challenges of Explainability and Traceability” (2024) 4 AI Ethics 229–245. 26 UN General Assembly, Report of the Secretary-General on Lethal autonomous weapons systems, A/79/88 (1 July 2024).

International law fully governs lethal autonomous weapons systems, encompassing a range of applicable legal frameworks. These include the United Nations Charter, customary international law, and key branches such as international humanitarian law, international human rights law, and international criminal law. Additionally, the principles of state responsibility further regulate the use and accountability surrounding such weapons. Together, these legal instruments ensure that autonomous weapons remain subject to established international legal standards. 27 Lethal autonomous weapons systems present difficulties in adhering to international legal frameworks, especially those concerning humanitarian law, human rights law, and criminal law. It was observed that no existing international treaty explicitly governs or bans such weapons.

28 Case Studies & Emerging State Practice Israel’s use of AI in Gaza - Research by Human Rights Watch suggests that certain technologies employed by Israeli forces depend on continuous and widespread surveillance of Gaza’s Palestinian population, utilizing data gathered even before the latest conflict in ways that violate international human rights standards. These systems analyze Palestinians' personal information to guide military decisions, such as assessing potential threats and selecting targets. Some of these tools incorporate machine learning—a form of artificial intelligence that identifies patterns and makes predictions without direct human programming. 29 Recent investigative reports30 reveal that Israel's AI-powered targeting platforms, notably "Lavender"31 and "The Gospel," are facilitating widespread civilian casualties and devastation throughout Gaza.

Following the October 7, 2023, Hamas-led attack in southern Israel, reports have detailed the Israeli military’s increasing reliance on digital technologies, including AI-driven surveillance systems, to guide airstrikes and operations in Gaza. 32 This raises serious concerns under

27 Ibid.

28 Ibid.

Questions and Answers: Israeli Military use of Digital tools in Gaza, available at: https://www.hrw.org/news/2024/09/10/questions-and-answers-israeli-militarys-use-digital-tools-gaza (last Visited on April 01, 2025)

30 A mass assassination factory: Inside Israel’s calculated bombing of Gaza, available at:

https://www.972mag.com/mass-assassination-factory-israel-calculated-bombing-gaza/ (last visited on April 01, 2025). Lavender: The

AI

machine directing Israel’s bombing spree in Gaza, available at: https://www.972mag.com/lavender-ai-israeli-army-gaza/ (last visited on April 01, 2025).

32 Ibid at 25

international humanitarian law, particularly regarding the principles of distinguishing combatants from civilians and taking all possible measures to avoid civilian casualties. The organization warns that the use of these systems may lead to violations of the laws of war, as they appear to operate on broad, pre-existing surveillance data collected unlawfully under human rights law. 33 NATO AI-guided targeting systems - The integration of AI-guided targeting systems by the U. S. and NATO presents significant advancements in military operations but also raises critical challenges to International Humanitarian Law (IHL). In response to the growing threat posed by Russian glide bombs, NATO has launched an innovation challenge to develop advanced countermeasures.

34 Among the promising solutions supported by NATO's innovation ecosystem is an AI-driven system developed by France's Alta Ares, capable of detecting, identifying, and predicting the trajectory of these bombs. Additionally, Germany's Tytan air drone offers an effective interception method by targeting and destroying glide bombs midflight. Another notable innovation comes from French startup Ateryd, which proposes a "drone wall" concept designed to intercept incoming glide bombs in mid-air—a solution considered highly relevant to Ukraine's current battlefield needs. 35 These cutting-edge technologies demonstrate NATO's commitment to leveraging innovation to address emerging battlefield threats.

This shows the aspiring desire of state as well as non-state actors to implement AI and Machine learning in their warfare technology, lacking proper laws to tackle with the attribution of responsibility in case of war crime committed by implementation of LAWS. Autonomous weapons programs in Russia - Ukraine conflict - Russian armed forces have deployed AI-enabled unmanned ground vehicles (UGVs) capable of autonomous target detection and engagement. These AI systems perform multiple battlefield functions, ranging from intelligence gathering to direct combat operations. Equipped with advanced sensor arrays, optical systems, and weapon mounts including automatic firearms, the UGVs operate independently to identify and neutralize threats without requiring human input during mission execution. 36 This development represents a significant advancement in automated warfare

33 Ibid at 25

34 NATO Selects AI and Drone Swarm Solutions That Effectively Counter Russian Glide Bombs, available at: https://united24media.com/latest-news/nato-selects-ai-and-drone-swarm-solutions-that-effectively-counterrussian-glide-bombs-7256 (last visited on April 01, 2025).

35 Ibid.

36 Kostenko, O., Jaynes, T., Zhuravlov, D., Dniprov, O. & Usenko, Y., “Problems of Using Autonomous Military AI Against the Background of Russia's Military Aggression Against Ukraine” (2022) (4) Baltic Journal of Legal and Social Sciences 131–145.

capabilities within contemporary conflict zones. Ukraine's armed forces have been actively creating and implementing autonomous combat systems, including the Phantom AI-powered turret. This defensive weapon system employs artificial intelligence for target recognition and engagement, functioning independently of direct human control. Equipped with smart sensors and visual recognition technology, the turret can distinguish between different target categories - from enemy soldiers to military vehicles - and automatically neutralize threats when detected. The automated defense platform represents Ukraine's technological adaptation to modern warfare demands, providing continuous protection for strategic locations without requiring constant human oversight.

37 The ongoing conflict has witnessed another significant development in autonomous warfare: Russia's deployment of AI-driven drone swarms. 38 These coordinated groups of unmanned aerial vehicles utilize artificial intelligence to operate independently, performing complex missions without direct human control. Each drone in the swarm carries sophisticated detection equipment, including advanced sensors and highresolution cameras, enabling precise target identification. Some units within these swarms are weaponized, capable of delivering explosive payloads or other munitions to designated targets. 39 This tactical approach represents a major evolution in autonomous combat systems, demonstrating how AI can enable collective decision-making among multiple robotic platforms in battlefield scenarios.

Conclusion The rapid development and deployment of AI-guided targeting systems and autonomous weapons in modern warfare present both unprecedented capabilities and profound ethical dilemmas. While these technological advancements offer military advantages in precision and efficiency, they simultaneously test the fundamental principles of International Humanitarian Law (IHL)—particularly distinction, proportionality, and precaution. The current legal framework, though robust, was not designed to address the complexities of AI-driven warfare. This gap highlights the urgent need for IHL to evolve proactively, rather than reactively, to 37 Onderco, M. , “Navigating the AI Frontier: Insights from the Ukraine Conflict for NATO’s Governance Role in Military AI” (2025) Journal of Strategic Studies 1–25.

Inside Russia’s plans to build autonomous drone swarms, available at: https://breakingdefense. com/2025/01/inside-russias-plan-to-build-autonomous-drone-swarms/ (last visited on April 01, 2025).

39 Ivan Nafuye, Weaponizing Artificial Intelligence in the Russia Ukraine war, available at:

https://www.researchgate.net/publication/379345343_Weaponizing_Artificial_Intelligence_in_the_Russia- Ukraine_war (last visited on April 01, 2025).

ensure it remains effective in regulating emerging technologies before they become entrenched in conflict. The international community faces a critical challenge: balancing military innovation with humanitarian obligations. Autonomous systems, from drone swarms to AIpowered turrets, demand clear legal boundaries to prevent unintended escalation and protect civilian lives. However, regulation alone is insufficient—there must also be transparency in development, accountability in deployment, and international cooperation to establish norms. As warfare becomes increasingly automated, the question is no longer just about what machines can do, but what they should do. The future of conflict hinges on whether humanity can uphold the values of IHL while navigating the age of autonomous weapons.

The time to act is now—by strengthening legal frameworks, fostering multilateral dialogue, and embedding ethical safeguards into the design of military AI. Only by confronting these challenges head-on can we ensure that technological progress does not outpace humanity’s commitment to just and lawful warfare.