India, Autonomous Weapons, And The Effects- Based Premise Of International Humanitarian Law
Author: LawSphereX
Category: International Law
Published:
Updated:
INDIA, AUTONOMOUS WEAPONS, AND THE EFFECTS-
BASED PREMISE OF INTERNATIONAL HUMANITARIAN LAW
Varada Arora, O.P. Jindal Global University
ABSTRACT
As States approach the Seventh Review Conference of the Convention on Certain Conventional Weapons in 2026, the question of whether to negotiate binding rules for lethal autonomous weapon systems remains open, and India occupies a striking position within it. Having chaired the Group of Governmental Experts that produced the first internationally agreed guiding principles on such weapons, India nonetheless voted against United Nations General Assembly Resolution 78/241. This essay argues that the doctrine underlying India's posture, that international humanitarian law regulates the effects of weapons rather than the technologies that produce them, so that existing law suffices, is internally coherent but unstable.
The effects-based premise governs autonomy well, where a system merely executes a human judgement already made, but falls silent where the system performs the legal reasoning of distinction and proportionality itself. The instability is sharpest in the law of individual responsibility, which presupposes a human author whom autonomy displaces: the Kargu-2 deployment in Libya shows that the author dissolves in practice. India's stance is strategically rational, rooted in a consistent refusal of discriminatory arms-control bargains. That strategic logic explains why India wants the existing law to suffice without showing that it does.
I.
Introduction
In October 2023, United Nations Secretary-General António Guterres and International Committee of the Red Cross President Mirjana Spoljaric urged States to conclude negotiations on a legally binding instrument governing autonomous weapon systems by 2026. Consequently, the year 2026 has emerged as a critical benchmark in international efforts to regulate autonomous warfare. 1 The Seventh Review Conference of the Convention on Certain Conventional Weapons (CCW), 2 will be held in November at the Palais des Nations in Geneva, Switzerland. The upcoming conference will reveal the key question of whether states will agree to start negotiating a new protocol or they will let the moment pass.
If they let it pass, the forum best placed to produce binding rules will lose its momentum, and the existing law of armed conflict will be left as the only check on weapons that can choose and strike targets on their own. 3 To understand the legal stakes of this choice, we must look at UNGA’s first resolution devoted to this particular matter. Resolution 78/241 reaffirmed that International law and International Humanitarian Law (IHL) in particular apply fully to lethal autonomous weapons systems (LAWS), and tasked the Secretary-General with gathering the views of states on how to address the humanitarian, legal, and ethical concerns these weapons raise. 4 India's opposition to Resolution 78/241 is particularly noteworthy. Only four states, including Belarus, Mali, Russia, and India, voted against the resolution.
5 Yet India had previously chaired the Group of Governmental Experts process that produced the first internationally agreed guiding principles on autonomous weapons systems. 6 This essay 1 International Committee of the Red Cross & United Nations Secretary-General, Joint Call for States to Establish New Prohibitions and Restrictions on Autonomous Weapon Systems (Oct. 5, 2023), https://www. icrc. org/en/document/joint-call-un-and-icrc-establish-prohibitions-and-restrictionsautonomousweapons-systems
2 Convention on Prohibitions or Restrictions on the Use of Certain Conventional Weapons Which May Be
Deemed to Be Excessively Injurious or to Have Indiscriminate Effects, opened for signature Apr. 10, 1981,
1342 U.N.T.S. 137.
3 U.N. Secretary-General, Lethal Autonomous Weapon Systems, U.N. Doc. A/79/88 (2024).
4 G.A. Res. 78/241 (Dec. 22, 2023).
5 Tejas Bharadwaj & Charukeshi Bhatt, India's Normative Stance on Lethal Autonomous Weapon Systems,
CARNEGIE ENDOWMENT FOR INT'L PEACE (Feb. 26, 2024), https://carnegieendowment.org/posts/2024/02/indias-normative-stance-on-lethal-autonomousweaponssystems.
6 Charukeshi Bhatt & Tejas Bharadwaj, Understanding the Global Debate on Lethal Autonomous Weapons
Systems: An Indian Perspective, CARNEGIE ENDOWMENT FOR INT'L PEACE (Aug. 2024), https://carnegieendowment.org/research/2024/08/understanding-the-global-debate-on-lethalautonomousweapons-systems-an-indian-perspective argues that the doctrine underlying India's posture, that IHL regulates the effects of weapons rather than the technologies that produce them, so that existing law suffices, is internally coherent but unstable. The cardinal principles of IHL, and the regime of individual responsibility that enforces them, are addressed to a human author of decisions whose displacement by autonomy strains the law from within. India thus defends a legal status quo whose underlying assumptions are increasingly challenged by the very forms of autonomy it seeks to accommodate. II.
The Effects-Based Premise: To see where this contested strain originates, one must take the effects-based premise seriously on its own terms, for within its proper domain, it could be considered correct. International Humanitarian Law has never regulated the internal mechanics of a weapon; it regulates what the weapon does. A munition that strikes only lawful targets and causes no disproportionate harm is lawful, whatever guides it to its mark, and a munition that cannot meet those conditions is unlawful, however it is controlled. Based on this premise, it can be argued that autonomy is merely a new means of delivery, and the existing principles of distinction, proportionality, and precaution are equal to the task of governing it. The premise holds wherever an autonomous system functions as a sophisticated instrument executing a human judgement already made.
The difficulty is that the most consequential autonomous systems are not built to execute a prior judgement but to make the judgement themselves. Distinction is not a fixed output but an assessment, of intent, of conduct, of whether a person has rendered themselves hors de combat. 7 The principle of proportionality requires weighing anticipated military advantage against civilian harm in circumstances the law assumes a human will read. When a system selects and engages targets without a contemporaneous human appraisal, it does not merely deliver an effect; it performs the legal reasoning the principles were written to discipline. The effectsbased account does not so much fail here as fall silent: it tells us when an effect is unlawful, but not who has assessed it, and that omission is precisely what autonomy forces into view.
The key principles, of IHL such as the principles of distinction and proportionality, do not enforce themselves. Their breach is answered, under the law of armed conflict, through individual criminal responsibility. Where a subordinate commits a violation, responsibility may 7 Protocol Additional to the Geneva Conventions of 12 Aug. 1949 and Relating to the Protection of Victims of International Armed Conflicts arts. 41, 48, June 8, 1977, 1125 U. N. T. S. 3 (hereinafter Additional Protocol I). ascend the chain of command. Under Article 86(2) of Additional Protocol I, 8and Article 28 of the Rome Statute, superiors may incur liability where they knew, or should have known, that subordinates were committing or about to commit violations and failed to take all feasible measures within their power to prevent, repress, or submit the matter of investigation and prosecution.
9 The doctrine is built around a foreseeable subordinate and a superior who could have intervened. III.
The Accountability Gap: An autonomous system fits neither role. It is not subordinate in any sense that the law can use; it bears no criminal responsibility of its own, possesses no mens rea, and is not a person against whom a predicate offence could be charged. Nor does liability settle comfortably on the human who deploys such a system. The commander who authorises a lethal autonomous weapon makes the decision to release it into an environment, the specific selection and engagement of a target and the very judgment is made afterwards, by the machine. The "knew or should have known" standard strains across that gap.
This is the responsibility gap Robert Sparrow identified: the programmer did not make the decision, the commander could neither predict nor control it, and the machine cannot be punished, so that holding any of them responsible for the particular wrong appears unjust, while holding no one responsible is intolerable. 10 The standard rejoinder is that a commander who unleashes an unpredictable weapon into a civilian setting assumes the risk, and that foreseeable indiscriminate harm is itself a form of recklessness. This has real force, but it does not close the gap rather, it relocates it. It secures accountability for the decision to deploy, not for the decision to kill; and where it fails, the fallback is state responsibility, a matter of inter-state reparation rather than the individual criminal sanction on which IHL's enforcement of grave breaches is built.
11 To accept that backstop is already to concede that the individual mechanism has frayed. 8 Protocol Additional to the Geneva Conventions of 12 August 1949 and Relating to the Protection of Victims of International Armed Conflicts (Protocol I), adopted June 8, 1977, 1125 U. N. T. S. 3. 9 Rome Statute of the International Criminal Court, opened for signature July 17, 1998, 2187 U. N. T. S. 3.
10 Robert Sparrow, Killer Robots, 24 J. APPLIED PHIL. 62, 66–69 (2007).
11 Draft Articles on Responsibility of States for Internationally Wrongful Acts arts. 1, 31, in Rep. of the Int'l Law Comm'n on the Work of Its Fifty-Third Session, U.N. Doc. A/56/10, at 26 (2001); Additional Protocol I, supra note 6, art. 91.
This does not remain merely theoretical. In March 2021, a UN Panel of Experts on Libya reported that retreating forces had been "hunted down" by STM Kargu-2 loitering munitions, which were "programmed to attack targets without requiring data connectivity between the operator and the munition: in effect, a true 'fire, forget and find' capability. "12 The episode is often described as the first autonomous killing of a human being, but that is precisely what the record does not establish: the report does not confirm that anyone was killed by a system operating without human supervision, nor that the munitions in fact engaged in autonomous mode at the decisive moment. 13 That ambiguity is the point.
Where the engagement decision may have been the machine's, the record is too thin to reconstruct who decided what, and a responsibility regime built on a foreseeable subordinate and a supervising superior cannot attach to a decision no one can be shown to have made. The accountability gap is not only doctrinal but evidentiary: the fog closes over the event before any human author can be named. The first real-world deployment thus stages the very problem the law has not resolved. It is in this context that India’s position is exposed. The Libyan episode is not a distant or isolated hypothetical that India dismisses; it is the form the accountability gap takes in the world, and India has committed on both sides of it.
India did not merely vote against Resolution 78/241; it had earlier chaired GGE, whose guiding principles affirm that human responsibility for decisions on the use of force must be retained. 14 India’s preferred remedy is not a binding rule but a voluntary political declaration that would, in its own framing, signal a high-level commitment to IHL and to human accountability — a declaration consciously chosen over legal constraint so as not to foreclose technological development. 15 That commitment names the right destination, and Kargu-2 is precisely the case it was written for: a deployment in which, if the engagement decision was the machine's, no human author can now be located to bear it. But to signal a commitment to accountability is to name the destination, not to build the road to it.
The effects-based premise India defends holds that existing IHL suffices because it governs what a weapon does, not how it does it. 12 U. N. Security Council, Letter dated 8 March 2021 from the Panel of Experts on Libya established pursuant to resolution 1973 (2011) addressed to the President of the Security Council, U. N. Doc. S/2021/229 (Mar. 8, 2021). 13 Hitoshi Nasu, The Kargu-2 Autonomous Attack Drone: Legal & Ethical Dimensions, Articles of War (June 10, 2021), https://lieber. westpoint. edu/kargu-2-autonomous-attack-drone-legal-ethical/ 14 Rep. of the 2019 Session of the Group of Governmental Experts on Emerging Technologies in the Area of Lethal Autonomous Weapons Systems, annex IV, ¶ (b), U. N. Doc. CCW/GGE. 1/2019/3 (Sept. 25, 2019).
15 Meghna Pradhar, India and the Global LAWS Debates, MANOHAR PARRIKAR INST. FOR DEF. STUD. &
ANALYSES (May 2025), https://idsa.in/publisher/issuebrief/india-and-the-global-laws-debates.
Accountability, however, is not a rule about effects; it is a rule about authorship. Libya shows the author dissolving at the moment the rule most needs one. India thus asserts the requirement — a human who answers for the killing, that its confidence in the sufficiency of existing law, and its preference for signalled commitment over binding doctrine, together leave unsecured. The principle India helped write names the gap; the case shows it opening; the position India has taken declines to close it. IV.
The Strategic Logic of the Paradox : However, one must consider India’s geopolitical strategy. India's posture is not carelessness dressed as principle; it is the considered product of a strategic environment, and taking it at its most rational is the only fair way to test it. India sits in a neighbourhood of two nucleararmed adversaries, and watches China invest heavily in "intelligentised" warfare, the infusion of autonomy and human-machine interface into weapon systems. In that setting, India cannot afford to fall behind in military AI, and still less can it afford to be locked into a legal framework that freezes a technological hierarchy favouring more advanced powers.
16 This is the familiar Indian objection to a discriminatory bargain, and India's treaty record is consistent: it declined the Nuclear Non-Proliferation Treaty 17and the Comprehensive Test Ban Treaty 18as asymmetric instruments that would disarm the weak while leaving the strong armed, and stayed out of the Ottawa Convention on anti-personnel mines for reasons of border security,19 while ratifying the Chemical and Biological Weapons Conventions,20 whose bans are universal and verified. India objects not to arms control as such, but to arms control that binds its hands while leaving rivals free. Read this way, the LAWS position is coherent. A binding prohibition negotiated now would, on India's view, be premature.
Locking in constraints around a technology still nascent and dual-use before states even agree on what the technology is, and would risk functioning as a technological chokehold imposed by those already ahead. Hence, India's preference for the CCW as the forum, the GGE's accumulated work as the foundation, and a voluntary political
16 Id. at 4.
17 Treaty on the Non-Proliferation of Nuclear Weapons, opened for signature July 1, 1968, 729 U.N.T.S. 161.
18 Comprehensive Nuclear-Test-Ban Treaty, opened for signature Sept. 24, 1996, 35 I.L.M. 1439.
19 Convention on the Prohibition of the Use, Stockpiling, Production and Transfer of Anti-Personnel Mines and on Their Destruction, opened for signature Sept. 18, 1997, 2056 U. N. T. S. 211. 20 Convention on the Prohibition of the Development, Production, Stockpiling and Use of Chemical Weapons and on Their Destruction, opened for signature Jan. 13, 1993, 1974 U. N. T. S. 45; Convention on the Prohibition of the Development, Production and Stockpiling of Bacteriological (Biological) and Toxin Weapons and on Their Destruction, opened for signature Apr. 10, 1972, 1015 U. N. T. S. 163. declaration over a treaty.
It is also why India has stayed out of the US-led Political Declaration on Responsible Military Use of Artificial Intelligence and Autonomy: not because it rejects that instrument's human-accountability language, which it shares, but because India will not let norm-setting migrate to a forum it does not control and cannot shape to its security needs. 21 But strategic coherence is not the same as legal sufficiency, and this is where the two strands of the paper meet. Every move in India's logic is a reason to keep the law flexible — to resist binding rules, preserve manoeuvre, avoid premature constraint. None of it answers the prior question the accountability gap raises: whether the existing law India insists is adequate can, in fact, reach the human it presumes. The strategic case explains why India wants the status quo to suffice; it does nothing to show that it does.
The paradox is therefore not resolved by understanding its motives. It is sharpened: India has the strongest of reasons to defend a legal order whose internal assumptions its own preferred technologies are unsettling. V.
Conclusion
We must note that AI picks up inadvertent biases from underlying data sets that are used to train it. “In the context of autonomous weapons, the criteria will inform who is and who is not a combatant or target will likely involve factors including gender, race, age, and inability” 22 The potential of artificial intelligence to revolutionise robotic drone swarms presents another particularly dangerous development. With the help of AI, these low-cost robots could form highly autonomous swarms capable of striking multiple targets simultaneously on a large scale, possibly challenging the principles of proportionality and precaution under IHL. 23 These dangers are not speculative refinements of a familiar weapon but a change in kind.
The ICRC has warned that a degree of unpredictability is inherent in autonomous weapons, since the user does not choose or know the specific target, timing, or location of the force applied, and that systems whose machine-learning functioning is opaque, the so-called "black-box" 21 U. S. Dep’t of State, Political Declaration on Responsible Military Use of Artificial Intelligence and Autonomy (Feb. 16, 2023), https://www. state. gov/political-declaration-on-responsible-military-use-of-artificialintelligenceand-autonomy.
22 Addressing Bias in Autonomous Weapons, U.N. Doc. CCW/GGE.1/2024/WP.5 (Mar. 8, 2024),
https://docslibrary.unoda.org/Convention_on_Certain_Conventional_Weapons_- Group_of_Governmental_Experts_on_Lethal_Autonomous_Weapons_Systems_(2024)/CCW-GGE.1- 2024WP.5.pdf
23 UN Addresses AI and the Dangers of Lethal Autonomous Weapons Systems, UNITED NATIONS
REGIONAL INFORMATION CENTRE FOR WESTERN EUROPE (May 17, 2024), https://unric. org/en/unaddresses-aiand-the-dangers-of-lethal-autonomous-weapons-systems/. problem, should, for that reason, be ruled out. 24 Swarming autonomy compounds the difficulty, multiplying these unscrutinised engagements faster than any human appraisal can track. India's resistance to binding rules is, as shown, strategically intelligible: it reflects a consistent refusal to be bound asymmetrically while rivals modernise unchecked. But a posture calibrated to preserve national manoeuvre is poorly suited to a problem that is structural rather than national. The accountability gap does not open because India, or any single state, deploys these systems; it opens because the law's enforcement presumes a human author that autonomy removes.
No voluntary declaration, however sincere, repairs an omission lodged in the architecture of the law itself. What this calls for is narrower than a ban and more than a declaration. A future instrument need not prohibit autonomous weapons outright; it could instead codify a requirement of meaningful human control over the selection and engagement of targets, and establish attribution rules that fix responsibility for an autonomous targeting decision before the system is deployed rather than after the harm is done. Such measures would not freeze the technology India is right to want; they would restore the human author the law already presumes. That is why the 2026 moment matters. If States let it pass, they leave the gravest decision in warfare, who shall be killed, governed by a body of law that can no longer name who decided.
The effects-based premise India defends answers the wrong question; the question autonomy forces, and that a new instrument would have to confront, is not what the weapon does, but who answers for it. 24 Int'l Comm. of the Red Cross, ICRC Position on Autonomous Weapon Systems and Background Paper (May 12, 2021), https://international-review. icrc. org/articles/icrc-position-on-autonomous-weapon-systemsicrcposition-and-background-paper-915.