217
© The Author(s) 2024
Scientific Article
DOI: 10.51940/2024.1.217-253
UDC: 341.33/.34:341.232:004.8
Maruša T. Veber*
Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
Abstract
The author analyses the notion of State consent in the delivery of humanitarian assis-
tance supported by artificial intelligence (AI) systems from the perspective of the existing 
applicable international legal regimes, in particular, the general legal regime of human-
itarian assistance and the specific rules deriving from international humanitarian law 
and international human rights law. She argues that the notion of consent lies at the 
heart of these rules with a distinction made between strategic and operational consent to 
humanitarian assistance. The former refers to a State’s general consent to the delivery of 
humanitarian assistance on its territory, while the latter refers to the consent required at 
the operational level for the delivery of a particular type of humanitarian assistance in a 
specific geographically defined area. It is argued that valid reasons for withholding opera-
tional consent to AI-supported humanitarian assistance under international humanitari-
an law must be distinguished from the arbitrary withholding of strategic consent. While 
withholding operational consent may be legally justified, the arbitrary withholding of 
strategic consent to humanitarian assistance is prohibited under the relevant interna-
tional legal regimes when it amounts to a violation of other existing obligations of the 
State concerned (e.g., under international humanitarian law or human rights law). In 
such situations the non-consensual delivery of humanitarian assistance could be legally 
justified either through United Nations Security Council authorisation or by secondary 
rules of international law, in particular countermeasures.
Key words
artificial Intelligence, humanitarian assistance, consent, arbitrary withholding of con-
sent, countermeasures.
* PhD, Assistant Professor and Associate Researcher, Department of International Law, Faculty of Law, 
University of Ljubljana, marusa.veber@pf.uni-lj.si.
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review  – voL. LXXXiv, 2024  •  pp. 217–253
ISSN 1854-3839  •  eISSN: 2464-0077
218
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review, voL. LXXXiv, 2024
1. Introduction**
Humanitarian assistance is increasingly being carried out by relying on digital infor-
mation technologies, including artificial intelligence (AI). AI systems,1 which typically 
draw on large amounts of data2, including the biometric data of aid recipients, have the 
potential to significantly enhance the accuracy and effectiveness of aid delivery, while 
also helping to prevent the misuse of humanitarian aid. By making the distribution of aid 
conditional on the use of AI and biometric data, the organisations mandated to deliver 
aid in the aftermath of man-made or natural disasters aim to ensure that the assistance 
reaches those in need, thereby preventing it from being diverted and used for other pur-
poses. However, the use of AI in a humanitarian context raises numerous important legal 
questions, including whether the aid-receiving State consents to the use of AI systems 
in its territory, and whether it might withhold consent due to concerns over the poten-
tial dual use of the collected data and the security of that data. Indeed, there have been 
instances where parties engaged in armed conflict have declined AI-supported humani-
tarian assistance provided by international humanitarian organisations, as evidenced by 
the situation in Yemen.
This paper analyses the notion of State consent in the delivery of humanitarian assis-
tance supported by AI systems, viewed from the perspective of the applicable internation-
al legal regimes—particularly the general legal regime of humanitarian assistance, as well 
as specific rules derived from international humanitarian law and international human 
rights law. It argues that consent lies at the heart of the rules governing the provision of 
humanitarian assistance, with a distinction drawn between strategic and operational con-
sent. Strategic consent refers to a State’s general consent to the delivery of humanitarian 
assistance on its territory, while operational consent refers to the consent required at the 
operational level for delivering a particular type of humanitarian assistance in a specific 
geographic area. It is argued that valid reasons for withholding operational consent to 
AI-supported humanitarian assistance under international humanitarian law must be dis-
tinguished from the arbitrary withholding of strategic consent. While the withholding of 
operational consent may be legally justified, the arbitrary withholding of strategic consent 
** This paper was prepared in the framework of a research project ‘Development and use of artificial intelli-
gence in light of the negative and positive obligations of the state to guarantee the right to life (J5-3107)’, 
which is co-funded by the Slovenian Research Agency (ARIS). See also T. Veber, 2024; and T. Veber, 2025.
1 There is currently no uniform definition of the AI. Arguably, the most authoritative definition was 
provided for in the UNESC Recommendation (2021), whereby AI systems are understood as “systems 
which have the capacity to process data and information in a way that resembles intelligent behaviour, 
and typically includes aspects of reasoning, learning, perception, prediction, planning or control”. 
UNESCO (2021), § 2. It is acknowledged, however, that the definition of AI will have to be changed 
over time in accordance with the rapid technological developments. T. Veber, 2023, pp. 14–15.
2 Beduschi, 2022, pp. 1149–1169.
219
Maruša T. Veber – Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
to humanitarian assistance is prohibited under the relevant international legal regimes. 
When analysing the consequences of arbitrarily withheld consent, this paper concludes 
that such withholding of consent cannot automatically confer legality on the non-con-
sensual delivery of AI-supported assistance in such situations.3 Rather, it argues that in 
certain limited cases, where the arbitrary withholding of consent amounts to a violation 
of other existing obligations of the State (e.g., under international humanitarian law or 
international human rights law), the non-consensual delivery of humanitarian aid could 
be legally justified either by United Nations Security Council (UNSC) authorisation or 
by existing secondary rules of international law, in particular the law of countermeasures.
It is acknowledged that, apart from the issue of State consent, the question of the con-
sent of individuals—namely the consent of recipients of humanitarian aid delivered with 
the support of AI—also arises in this context. Specifically, the use of AI in humanitarian 
assistance raises various questions relating to data protection and the right to privacy 
of the individuals concerned.4 However, an analysis of this topic lies beyond the scope 
of this paper. It also must be noted that this paper focuses solely on the delivery of aid 
by international organisations, such as UN specialised agencies, rather than the work of 
non-governmental organisations (NGOs) mandated with delivering aid. Since NGOs are 
not subjects of international law stricto sensu, they are primarily governed by national laws, 
and the application of international legal rules applies differently to them than to inter-
national organisations.5 Finally, this paper focuses on the notion of consent in the context 
of AI-supported humanitarian assistance, a concept specific to the broader question of 
the possible non-consensual provision of humanitarian assistance under international law.
Following this introduction, the paper briefly presents the practice of using AI by 
international humanitarian organisations (section 2). Section 3 discusses the AI-specific 
legal regimes, while section 4 analyses relevant international legal regimes governing the 
provision of humanitarian assistance under international law. The paper then outlines 
the modalities and the distinction between withholding operational consent and with-
holding strategic consent to humanitarian assistance (section 5) and discusses the (il)
legality of the non-consensual provision of humanitarian assistance under international 
law (section 6). The final section (7) presents possible legal justifications for the non-con-
sensual delivery of humanitarian assistance under the UNSC collective security regime 
and the secondary rules on responsibility, with special emphasis on the law of counter-
measures. Finally, section 8 offers concluding remarks.
3 This argument is put forward for example by Barber, 2023.
4 More on this see: T. Veber, 2025. See also Narbel & Sukaitis, 2021; European Data Protection 
Board, 2022, p. 10; FRA, 2020; Wills, 2019; Kuner & Marelli, 2020, pp. 280–296.
5 Kuner, 2020, p. 81; Generally, on non-governmental organizations see: Lindblom, 2009.
220
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review, voL. LXXXiv, 2024
2. AI-Supported Humanitarian Assistance
Humanitarian assistance is increasingly provided through reliance on digital infor-
mation technologies, including AI. For example, the World Food Programme (WFP), 
in partnership with the United Nations High Commissioner for Refugees (UNHCR), 
introduced an iris-scan payment system in a Jordanian refugee camp, enabling 76,000 
Syrian refugees to purchase food from camp supermarkets using only an iris scan instead 
of cash, vouchers, or e-cards.6 This system connects with different databases within sec-
onds (e.g., the UNHCR and bank databases), thereby enabling quick and efficient aid 
delivery. By making the distribution of aid conditional on the use of AI and biometric 
data, these organisations aim to ensure that the assistance goes directly to those in need, 
preventing its diversion for other purposes.7 However, the use of AI in a humanitarian 
context also raises numerous important legal questions.
One particular concern is that AI systems in humanitarian assistance may be prob-
lematic because of the possible dual use of the data these systems collect. AI systems run 
on a variety of datasets and produce large amounts of data, which may help improve 
aid delivery. Yet, that same data can easily be used for other purposes and become tools 
for surveillance, security checks, tracing, or deportation.8 The issue of data security and 
the potential compromises of sensitive data is particularly relevant since, in the past, 
cyberattacks on humanitarian organisations exposed the personal data of about 500,000 
vulnerable people around the world.9 Moreover, requests have been made by different 
States to access biometrics data on refugees from humanitarian organisations, to use such 
data for security checks and deportation procedures.10
In addition, international organisations mandated with, for example, food assistance 
are increasingly relying on private commercial actors to support their humanitarian ac-
tivities. One illustrative example is the WFP, which pledged to “become a digitally ena-
bled and data-driven organization, with investments in new technology”,11 and recently 
partnered with Palantir to use its software to provide faster and more efficient food as-
sistance.12 Palantir is a leading US company specialising in data analytics, which is also 
increasingly integrating AI into various aspects of its operation, ranking among the top 
AI software platforms.13 Palantir has, however, been the subject of criticism, with allega-
6 WFP, 2016.
7 Reuters, 2019.
8 Martin et al., 2023, pp. 1363–1397.
9 Macdonald, 2022.
10 In the past, Bangladesh, Lebanon, Malaysia, and the US, for example, requested access to UNHCR 
biometric data on refugees. Martin et al., 2023, p. 1382.
11 WFP strategic plan (2022-2023), WFP/EB.2/2021/4-A/1/Rev.2, 12 November 2021, § 130.
12 Parker, 2019.
13 Businesswire, 2022.
221
Maruša T. Veber – Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
tions that it provided controversial data-sifting software to US government agencies.14 In 
this context, the term “surveillance humanitarianism” is sometimes used to describe the 
potential widespread collection of data in a humanitarian context without adequate safe-
guards—an approach that may “inadvertently amplify the vulnerability of individuals in 
need of humanitarian aid”.15 Others refer to “techno-colonianism”, wherein practices of 
digital innovation “can lead to reproducing the colonial relationships of dependency and 
inequality amongst different populations around the world.”16
Due to these concerns, parties to an armed conflict have, on occasion, refused AI-
supported humanitarian aid. For example, in 2019, the WFP decided to suspend the 
delivery of food aid in Yemen because of a disagreement on about using technology that 
employed biometric data (via iris scans, fingerprints, or facial recognition) to support aid 
delivery to food recipients.17 The principal objection was the concern that utilising AI 
and collecting data could jeopardise the security of the State.18
3. The AI-Specific Legal Framework
In 2024 two legally binding AI-specific documents were adopted. The Council of 
Europe adopted the first international treaty regulating the development and use of AI 
systems, which stresses the need for the application of existing human rights obligations 
to the development and use of AI systems, and provides some concrete safeguards in 
this respect. The Framework Convention on Artificial Intelligence and Human Rights, 
Democracy and the Rule of Law (Framework Convention on AI)19 is based on the fol-
lowing fundamental principles: human dignity and individual autonomy, equality and 
non-discrimination, respect for privacy and personal data protection, transparency and 
oversight, accountability and responsibility, and reliability and safe innovation.20
On the other hand, at the European Union (EU) level, the AI Act21 was adopted, 
governing the development and use of AI. In terms of substantive provisions, the AI 
Act is based on the so-called ‘risk-based’ approach. This means that it categorises AI 
14 Martin et al., 2023, p. 1363; BBC, 2020.
15 Latonero, 2019, as cited in Beduschi, 2022, p. 1152.
16 Madianou, 2019, as cited in Beduschi, 2022, p. 1152.
17 Reuters, 2019; Welsh, 2019.
18 Martin et al., 2023, p. 1364.
19 Council of Europe Framework Convention on Artificial Intelligence and Human Rights, 
Democracy and the Rule of Law, Council of Europe Treaty Series – No. [225], 2024.
20 More on this see T. Veber, 2025.
21 Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying 
down harmonised rules on artificial intelligence and amending Regulations (EC) No 300/2008, 
(EU) No 167/2013, (EU) No 168/2013, (EU) 2018/858, (EU) 2018/1139 and (EU) 2019/2144 
222
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review, voL. LXXXiv, 2024
systems according to the level of risk they might pose from the perspective of health, 
safety, fundamental rights, the environment, democracy or the rule of law, into: prohib-
ited AI practices, high-risk systems listed in Annex III, general-purpose AI models with 
systemic risk, and general-purpose AI models. While AI systems with unacceptable risks 
are prohibited, high-risk systems are subject to certain requirements in terms of data 
quality22, transparency, 23 human oversight, 24 fundamental rights impact assessment25 
and registration.26 Under the AI Act certain biometric identification27 systems fall under 
prohibited practices, e.g. biometric categorisation systems that categorise individual nat-
ural persons28 and ‘real-time’ remote biometric identification systems in publicly acces-
sible spaces for the purposes of law enforcement (except in certain limited cases).29 On 
the other hand, uses of other types of biometrics (e.g., remote biometric identification 
systems) would have to comply with requirements for high-risk AI systems.30
While these documents regulate the development and deployment of AI systems, 
including biometric systems, their relevance for the present paper is limited for the fol-
lowing two reasons:
1. Humanitarian international organisations, such as WFP, are not parties to these tre-
aties and, even in cases where the use of AI systems by humanitarian international 
organisations would fall under the material and territorial scope of these laws, the 
enforcement of these rules is foreclosed by the privileges and immunities to which IOs 
are entitled under international law31;
2. These two documents regulate AI products within their member States/signatories 
and provide safeguards concerning the protection of the human rights of indivi-
duals possibly affected by the use of AI. They do not explicitly address the possible 
non-consensual use of AI in the territory of a country affected by a humanitarian 
catastrophe of a sort.
and Directives 2014/90/EU, (EU) 2016/797 and (EU) 2020/1828 (Artificial Intelligence Act), OJ 
L, 2024/1689, 12 July 2024 (AI Act).
22 AI Act, Article 10.
23 AI Act, Article 13.
24 AI Act, Article 14.
25 AI Act, Article 27.
26 AI Act, Article 49.
27 According to Article 3(35) AI Act ‘biometric identification’ means the automated recognition of 
physical, physiological, behavioural, or psychological human features for the purpose of establish-
ing the identity of a natural person by comparing biometric data of that individual to biometric 
data of individuals stored in a database.
28 AI Act, Article 5(1)(g).
29 AI Act, Article 5(1)(h).
30 AI Act, Section 2.
31 For a detail analysis of this see T. Veber, 2025. See also Kuner, 2019, p. 174.
223
Maruša T. Veber – Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
At the United Nations (UN) level, which is most relevant to our discussions, AI is 
mainly being addressed through soft-law documents. Arguably, the most important 
development in this respect is the adoption of the ‘Principles for the Ethical Use of 
Artificial Intelligence in the United Nations System’ by the Inter-Agency Working 
Group on Artificial Intelligence in 2022.32 These principles aim to guide the design, de-
velopment, deployment and use of AI by UN agencies through the following principles: 
do no harm; defined purpose, necessity and proportionality; safety and security; fairness 
and non-discrimination; sustainability; right to privacy, data protection and data gov-
ernance; human autonomy and oversight; transparency and explainability; responsibil-
ity and accountability; and inclusion and participation. While these principles provide 
valuable guidance for the use of AI by UN agencies, including those mandated with 
the provision of humanitarian assistance, their primary aim is to safeguard the rights 
of individuals subject to AI systems. In this respect, no explicit legal obligations can be 
derived from these principles for international organisations providing AI-supported 
humanitarian assistance in terms of the consent of the concerned, aid-receiving State, to 
the use of AI on its territory.
However, the foregoing does not mean that the delivery of AI-supported humanitar-
ian aid by humanitarian international organizations remains unregulated as a legal lacu-
na. Activities of international organisations utilising AI in their humanitarian delivery 
missions are governed by the existing applicable international legal regimes, which are 
analysed in the remaining part of this paper.
4. International Legal Regimes Governing Humanitarian Assistance and 
the Issue of Consent
Deriving from the principle of sovereignty and its corollary, the principle of non-in-
tervention,33 it is the primary responsibility of the affected State to ensure, organise, 
coordinate, and implement the protection of affected persons and provision of humani-
tarian assistance in cases of natural disasters and other emergencies occurring on its terri-
tory, or on the territory under its jurisdiction or control.34 When States possess adequate 
32 UN, Principles for the Ethical Use of Artificial Intelligence in the United Nations System, 20 
September 2022.
33 Articles 2(1) and 2(7) Charter of the United Nations (UN Charter) (24 June 1945, entered into 
force 24 October 1945) 1 UNTS XVI (UN Charter); Declaration on Principles on International 
Law Concerning Friendly Relations and Cooperation among States in accordance with the 
Charter of the United Nations, UNGA resolution 2625 (XXV), 24 October 1970, UN Doc. A/
RES/2625(XXV).
34 Draft Article 10, Draft articles on the protection of persons in the event of disasters, YILC 2016, vol. 
II Part Two; UNGA Resolution 46/182, 19 December 1991, UN Doc. 46/182, Annex, Guiding 
Principles, § 4. See also Institute of International Law, Resolution of Humanitarian Assistance, 
Bruges Session – 2003, 8 September 2003 (IIL Resolution 2003), p. 5.
224
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review, voL. LXXXiv, 2024
capacities to respond to man-made or other humanitarian disasters, the issues of consent 
will generally not arise. However, in cases where a disaster exceeds national response 
capacities, civilians in need are inadequately provided with essential supplies. The assis-
tance of other actors, including international organisations as part of the international 
community, is warranted through the provision of impartial humanitarian assistance to 
the affected State.35 The question that hereby arises is whether the concerned State has an 
obligation to accept such humanitarian assistance and whether humanitarian assistance 
could potentially be provided in the absence of the affected State’s consent. In the context 
of AI-supported humanitarian assistance, the use of AI technology, such as iris scanning, 
may be the main reason for withholding consent.
Three relevant legal regimes applicable to situations of the delivery of humanitar-
ian assistance in cases of man-made or natural disasters are: the general humanitarian 
assistance legal regime, the international human rights regime and the international hu-
manitarian law regime. The first two will generally apply to humanitarian situations not 
involving an armed conflict. In times of armed conflict, however, a specific regime con-
cerning the provision of humanitarian assistance exists under international humanitarian 
law. Even though the general humanitarian assistance legal regime and international 
human rights law continue to apply in such situations, in times of war, the human rights 
regime is to be interpreted in light of the rules of international humanitarian law, which 
are applicable as lex specialis.36
4.1. General Humanitarian Assistance Legal Regime
Humanitarian assistance in cases of both natural37 and man-made38 disasters and 
emergencies has continuously occupied the agenda of the UN, with the UN General 
Assembly (UNGA) Resolution 46/182 (1991)39 outlining the guiding principles of hu-
manitarian assistance as the cornerstone of this regime. Among the key guiding principles 
embedded in this document are the principles of humanity, neutrality and impartiality. 
35 Oxford Guidance on the Law Relating to Humanitarian Relief Operations in Situations of Armed 
Conflict, 2016, § 6.
36 Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territory, Advisory 
Opinion (2004) ICJ Rep. 136, §§ 106–113.
37 UNGA Resolution 2034 (XX), 7 December 1965, UN Doc. A/RES/2034; UNGA Resolution 
44/236, 22 December 1989, UN Doc. 44/236. The Institute of International Law, classified »di-
sasters« as either natural, man-made disasters of technological origin or disaster caused by armed 
conflicts or violence. IIL Resolution 2003.
38 The Institute of International Law, classified »disasters« as either natural, man-made disasters of 
technological origin or disaster caused by armed conflicts or violence. IIL Resolution 2003. UNGA 
Resolution 2816 (XXVI), 14 December 1971, UN Doc. A/RES/2816.
39 UNGA Resolution 46/182, 19 December 1991, UN Doc. 46/182.
225
Maruša T. Veber – Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
In terms of the consent of the affected State to external humanitarian assistance, the third 
guiding principle centres around the prior consent of the affected State:
“[T]he sovereignty, territorial integrity and national unity of States must be fully 
respected in accordance with the Charter of the United Nations. In this context, 
humanitarian assistance should be provided with the consent of the affected coun-
try and in principle on the basis of an appeal by the affected community.”40
The issue of consent has also been debated within the International Law Commission’s 
(ILC) work on the Draft articles on the protection of persons in the event of disasters, 
which explicitly acknowledge that “the provision of external assistance requires the con-
sent of the affected state.”41 Similarly, the notion of consent is central in the work of the 
Institute of International Law (IIL) on the matter, whereby its resolution on humanitar-
ian assistance stresses that “States and organizations have the right to provide humani-
tarian assistance to victims in the affected States, subject to the consent of these States.”42 
Accordingly, State sovereignty and the notion of consent seem to lie at the heart of the 
general humanitarian assistance regime.
Simultaneously, however, both the ILC and the IIL recognise that in cases of disas-
ters exceeding national response capacities, the concerned State “shall seek” 43 or “has a 
duty to seek”44 assistance from competent international organisations, third States and 
other actors. As explained in the commentaries, the ILC embedded this reasoning on the 
basis of the principle of sovereignty, which does not only confer rights upon States but 
also imposes certain duties.45 However, at the same time, the ILC, in the commentary, 
stressed that the term “seek” cannot be equated with a duty to give consent to humanitar-
ian assistance but rather “entails the proactive initiation by an affected State of a process 
through which agreement may be reached.”46 Consent to humanitarian assistance, there-
fore, remains central to this regime, with the ILC acknowledging that “the provision of 
40 UNGA Resolution 46/182, 19 December 1991, UN Doc. 46/182, Annex, Guiding Principles, § 
3. See also UNGA Resolution 67/87, UN Doc. A/RES/67/87, 26 March 2013 (“Emphasizing also 
the fundamentally civilian character of humanitarian assistance, and, in situations in which military 
capacity and assets are used to support the implementation of humanitarian assistance, reaffirming 
the need for their use to be undertaken with the consent of the affected State and in conformity with 
international law, including international humanitarian law, as well as humanitarian principles.”).
41 Draft articles on the protection of persons in the event of disasters (2016), Article 13.
42 IIL Resolution 2003, § IV(2).
43 IIL Resolution 2003, § III(3).
44 Draft articles on the protection of persons in the event of disasters (2016), Article 11.
45 Draft articles on the protection of persons in the event of disasters (2016), Commentary to Article 
10, § 3.
46 Draft articles on the protection of persons in the event of disasters (2016), Commentary to Article 
11, § 6.
226
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review, voL. LXXXiv, 2024
external assistance requires the consent of the affected State [which] is fundamental to 
international law.”47
4.2. International Human Rights Law Regime
The obligation to provide humanitarian assistance to civilian populations in need 
also stems from the international human rights law regime, particularly from the pro-
visions on the right to life deriving from Article 6 of the International Covenant on 
Civil and Political Rights (ICCPR)48 and the right to food as enshrined in Article 11 of 
the International Covenant on Economic, Social and Cultural Rights (ICESCR)49. As 
already mentioned, it is generally confirmed that the obligations of States under inter-
national human rights law continue to apply in times of armed conflict50, including in 
times of occupation.51
The right to life under Article 6(1) ICCPR is non-derogable under the ICCPR52 even 
in “time of public emergency which threatens the life of the nation,” which includes situ-
ations of armed conflict and other public emergencies.53 The Human Rights Committee 
explained States’ obligations deriving from this provision in its General Comment 36, 
whereby it affirmed that the right to life “should not be interpreted narrowly” and in-
cludes “the entitlement of individuals to be free from acts and omissions that are intend-
ed or may be expected to cause their unnatural or premature death, as well as to enjoy 
a life with dignity.”54 Moreover, it expressly recognised that the positive obligations of 
States (the duty to protect life) include taking “appropriate measures to address the gen-
eral conditions in society that may give rise to direct threats to life or prevent individuals 
from enjoying their right to life with dignity”, including “widespread hunger and mal-
47 Draft articles on the protection of persons in the event of disasters (2016), Commentary to Article 
13, § 2.
48 International Covenant on Civil and Political Rights, 16 December 1966, UNTS, vol. 999, p. 171 
(ICCPR).
49 International Covenant on Economic, Social and Cultural Rights, 16 December 1966, UNTS, vol. 
993, p. 3 (ICESC).
50 Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territory, Advisory 
Opinion (2004) ICJ Rep. 136, §§ 106–113.
51 See, e.g., UN Human Right Committee (UN HRC), General Comment No. 31 UN Doc. 
CCPR/C/Rev.1/Add.13 (2004), § 10; Akande and Gillard, 2016, p. 504.
52 ICCPR, Article 4(2).
53 See also UN HRC, General Comment No. 29: Article 4: Derogations during a State of Emergency, 
31 August 2001, CCPR/C/21/Rev.1/Add.11.
54 UN HRC General comment no. 36, Article 6 (Right to Life), 3 September 2019,  CCPR/C/
GC/35, § 3 (General comment no. 36).
227
Maruša T. Veber – Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
nutrition and extreme poverty and homelessness.”55 In this respect, it confirmed that the 
right to life includes the obligation to ensure access to humanitarian assistance:
“The measures called for to address adequate conditions for protecting the right to 
life include, where necessary, measures designed to ensure access without delay by 
individuals to essential goods and services such as food, water, shelter, health care, 
electricity and sanitation, and other measures designed to promote and facilitate 
adequate general conditions, such as the bolstering of effective emergency health 
services, emergency response operations.”56
On the other hand, according to the International Covenant on Economic, Social and 
Cultural Rights, the right to food as embedded in Article 11 includes the obligation that:
“The States Parties will take appropriate steps to ensure the realization of this right, 
recognizing to this effect the essential importance of international co-operation 
based on free consent.”57
Unlike civil and political rights, economic, social and cultural rights may not be dero-
gated from in times of emergency, which is compensated by the fact that they are subject 
to progressive realisation, i.e. dependent on the available resources of States. Therefore, 
even in emergencies, States have to do their best to work towards the progressive realisa-
tion of these rights and guarantee the minimum content of the core obligations.58
The Committee on Economic, Social and Cultural Rights expressly stressed in General 
Comment No. 12 on the Right to Adequate Food that violations of the right to food:
“can occur through the direct action of States or other entities insufficiently regu-
lated by States. These include […] the prevention of access to humanitarian food 
aid in internal conflicts.”59
While indeed economic, social and cultural rights are subject to progressive realisa-
tion, States are under an obligation to ensure “minimum essential levels” of these rights.60 
In this respect, the Committee also stressed the need to seek international assistance to 
secure available resources for the realisation of the right to food. For a State not to be in 
breach of Article 11 by failing to ensure “at the very least, the minimum essential level 
required to be free from hunger” in cases of natural or man-made disasters, it has to 
55 UN HRC General comment no. 36, § 26.
56 Ibid.
57 ICESCR, Article 11(1).
58 Organization for Security and Co-operation in Europe Office for Democratic Institutions and 
Human Rights, Report on Violations of International Humanitarian and Human Rights Law, War 
Crimes And Crimes Against Humanity Committed in Ukraine (1 April – 25 June 2022), ODIHR.
GAL/36/22/Corr.1, 14 July 2022, p. 83.
59 UN Committee on Economic, Social and Cultural Rights (CESCR), General Comment No. 12: 
The Right to Adequate Food (Art. 11 of the Covenant), 12 May 1999, E/C.12/1999/5, § 19.
60 UN CESCR General Comment No. 3: The Nature of States Parties Obligations, UN Doc. 
E/1991/23, 14 December 1990, § 10.
228
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review, voL. LXXXiv, 2024
“demonstrate that every effort has been made to use all the resources at its disposal in an 
effort to satisfy, as a matter of priority, those minimum obligations,” including that it has 
sought to obtain international support.61
Moreover, according to the Special Representative of the UN Secretary-General on 
internally displaced persons, the obligation to allow for third-party provision of human-
itarian assistance also stems from other rights, such as the right to an adequate standard 
of living, health and education.62
Against this background, certain obligations concerning the provision of humani-
tarian assistance, including a duty to seek assistance from the international community, 
explicitly derive from international human rights law. To fulfil their international obliga-
tions towards individuals, States may, therefore, have to resort to international support in 
cases where their resources are inadequate to meet protection needs.63 If they fail to do so, 
they risk breaching their above-mentioned obligations under the ICCPR and ICESCR.
However, these obligations cannot be translated into a general obligation to give 
unconditional consent to the provision of humanitarian assistance on the territory of the 
concerned State to protect the right to life of its citizens and realise its progressive obliga-
tion under the right to food. As will be explained below, within the international human 
rights legal framework, the question of States’ human rights obligations and possible 
violations is to be determined against the background of the concrete circumstances of a 
situation, whereby the question of whether denial of consent to humanitarian assistance 
is necessary and proportionate to achieving legitimate ends is generally assessed in the 
context of arbitrariness.
4.3. International Humanitarian Law Regime
The obligation to provide humanitarian assistance to the civilian population in times 
of armed conflict is one of the central aspects of international humanitarian law. It is 
acknowledged that under this regime different modalities of humanitarian assistance 
arise in different contexts: international armed conflict, non-international armed con-
61 UN CESCR General Comment 12, 1999, § 17.
62 “A State is deemed to have violated the right to an adequate standard of living, to health and to 
education, if authorities knew or should have known about the humanitarian needs but failed to 
take measures to satisfy, at the very least, the most basic standards imposed by these rights. State 
obligations thus include the responsibility to follow up on these situations of concern and assess 
relevant needs in good faith, and ensure that humanitarian needs are being met, by the State itself or 
through available assistance by national or international humanitarian agencies and organizations, 
to the fullest extent possible under the circumstances and with the least possible delay.” Report of 
the Representative of the Secretary-General on the human rights of internally displaced persons, 
UN Doc. A /65/282, 11 August 2010, § 69.
63 Draft articles on the protection of persons in the event of disasters (2016), Commentary to Article 
11, § 3.
229
Maruša T. Veber – Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
flict, occupation, and provision of humanitarian assistance on territories controlled by 
non-state actors.64 Outlining all modalities of the relevant rules governing the provision 
of humanitarian assistance in these contexts is beyond the scope of this paper. Rather, 
this section merely clarifies the role of the consent of the concerned State in the provision 
of humanitarian assistance, which seems to lie at the centre of the international human-
itarian law regime.
At the outset, it has to be explained that under the international humanitarian law 
regime, two different levels of consent exist:
1. At the strategic level, humanitarian international organisations have to seek the con-
sent of the concerned State to enter the territory or territories in question (the so-cal-
led strategic consent); and
2. At the operational level, once this strategic consent has been obtained, the provision of 
humanitarian assistance is subject to the right of control by the parties to the conflict.65
In other words, parties to the conflict are to give operational consent to the provision 
of specific humanitarian aid in a certain geographic area and may prescribe technical ar-
rangements for the passage of such humanitarian assistance, search for humanitarian aid 
to verify the humanitarian nature of supplies, prevent convoys from affecting or being 
affected by military operations and ensure supplies meet health and safety standards.66 
While this section primarily addresses strategic consent, operational consent is analysed 
in the following section.
In terms of strategic consent under international humanitarian law, parties to the 
conflict are obliged to allow free passage of humanitarian assistance to those in need.67 
The International Committee of the Red Cross (ICRC) emphasised on numerous occa-
sions the importance of unimpeded access to humanitarian assistance by civilian popula-
tions in times of armed conflict, in accordance with the applicable rules of international 
humanitarian law.68 According to customary international humanitarian law applicable 
to international and non-international armed conflicts,
“[P]arties to the conflict must allow and facilitate unimpeded passage of huma-
nitarian relief for civilians in need, which is impartial in character and conducted 
without any adverse distinction subject to their right of control.”69
64 Ryngaert, 2013, pp. 6–9.
65 See also Sharpe, 2023.
66 Oxford Guidance on the Law Relating to Humanitarian Relief Operations in Situations of Armed 
Conflict, 2016, §§ 65–72.
67 Henckaerts, 2005, Rule 55.
68 International Humanitarian Law Databases, Customary IHL, Rule 55, <https://ihl-databases.icrc.
org/en/customary-ihl/v1/rule55#refFn_E763511D_00020> (accessed 30 April 2023).
69 Henckaerts, 2005, Rule 55.
230
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review, voL. LXXXiv, 2024
The provision of humanitarian assistance in international70 and non-international71 
armed conflicts is also governed by relevant treaty law. The Fourth Geneva Convention 
obliges States to allow for “the free passage of all consignments of essential foodstuffs,”72 
whereby Additional Protocol I broadens this obligation to the “rapid and unimpeded 
passage of all relief consignments, equipment and personnel.”73 In times of occupation, 
the occupying power “shall agree to relief schemes on behalf of the respective population 
and shall facilitate them by all the means at its disposal.”74 The obligation to allow for and 
facilitate access to humanitarian relief for civilians in need is also enshrined in national 
military manuals and is supported by State practice.75
The question that often arises in situations of armed conflict is whether there exists 
an obligation of parties to a conflict to give strategic consent to humanitarian assistance 
and whether assistance could be provided without such strategic consent. In this respect, 
relevant rules specifically regulating humanitarian assistance in non-international and 
international armed conflicts emphasise the central role of the consent of the affected 
State, whereby the obligation to allow for the free passage of humanitarian assistance is 
preconditioned by “consent of the High Contracting Party concerned,”76 or is “subject 
to the agreement of the Parties concerned in such relief actions.”77 The requirement of 
consent also clearly stems from Additional Protocol II, which preconditions delivery of 
humanitarian assistance in non-international armed conflicts with the explicit consent of 
the parties to the conflict. It has sometimes, therefore, been argued that, in relation to 
non-signatories of the AP II, delivery of aid in non-international armed conflicts could be 
non-consensual.78 However, the requirement of consent can also be implied from other 
provisions and customary international law, as it is hard to imagine how parties to a con-
70 Geneva Convention Relative to the Protection of Civilian Persons in Time of War (Fourth Geneva 
Convention), 12 August 1949,  75 UNTS 287, Articles 23 and 59; Protocol Additional to the 
Geneva Conventions of 12 August 1949, and relating to the Protection of Victims of International 
Armed Conflicts (Protocol I), 8 June 1977, 1125 UNTS 3, Articles 69–71.
71 Common Article 3(2) of the Geneva Conventions; Protocol Additional to the Geneva Conventions 
of 12 August 1949, and relating to the Protection of Victims of Non-International Armed Conflicts 
(Protocol II), 8 June 1977, 1125 UNTS 609, Article 18.
72 Fourth Geneva Convention, 1949, Article 23.
73 Protocol I, 1977, Article 70(2). See also Protocol II, 1977, Article 18(2).
74 Fourth Geneva Convention, 1949, Article 59. See also Fourth Geneva Convention, 1949, Article 62.
75 International Humanitarian Law Databases, Customary IHL, Rule 55, <https://ihl-databases.icrc.
org/en/customary-ihl/v1/rule55#refFn_E763511D_00020> (accessed 30 April 2023).
76 Protocol II, Article 18(2).
77 Protocol I, Article 70(1).
78 Barber, 2023, p. 2; Sproson & Olabi, 2023, p. 1 ff; see also: American Relief Coalition for Syria, 
2022, pp. 25–36.
231
Maruša T. Veber – Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
flict could make use of their right to “control”79 the provision of humanitarian assistance 
(operational consent) without previously giving strategic consent to such assistance.80
It is (probably) against this background that the ICRC concludes it is considered 
“self-evident” that a humanitarian organisation cannot “operate (in states) without the 
strategic consent of the party concerned,” both in international and non-international 
armed conflicts,81 and that some scholars talk about the “absolute” nature of the require-
ment of consent.82 This is not without problems and does not mean that States have no 
obligations concerning humanitarian assistance, nor that they can arbitrarily withhold 
consent. As will be explained in the following section, arbitrarily withholding consent 
to humanitarian assistance typically amounts to a violation of international law. It does, 
however, confirm that humanitarian organisations generally would not operate in affect-
ed States without their consent.
This seems to be endorsed by the UNSC, which in its resolutions often called for 
unimpeded access to humanitarian assistance in conflict situations,83 while at the same 
time also reaffirming the commitment of UN member States to respect the sovereignty, 
territorial integrity and political independence of the aid-receiving State,84 and urging all 
parties in a particular situation to facilitate the delivery of humanitarian assistance in ac-
cordance with international humanitarian law.85 The important role of strategic consent 
is also implied in relevant resolutions of the UNGA, whereby it called on affected States 
to facilitate the work of humanitarian organisations,86 not outlining, however, that they 
are legally obliged to do so unconditionally.
The question of consent under international humanitarian law is especially pertinent 
in situations whereby part of the territory of the conflict-affected State is controlled by 
non-state actors. In such situations, States are especially inclined, for military reasons, 
to deny humanitarian assistance in these areas, as was, for example, the case in Syria.87 
79 Henckaerts, 2005, Rule 55.
80 Similar conclusion is reached in the Oxford Guidance on the Law Relating to Humanitarian Relief 
Operations in Situations of Armed Conflict, 2016, § 30.
81 Henckaerts & Doswald-Beck, 2005, Commentary to Rule 55, pp. 195–200. See also Report of 
the Secretary-General on the protection of civilians in armed conflict, UN Doc. S /2013/689, 22 
November 2013, § 58.
82 Akande & Gillard, 2016, p. 489.
83 See, e.g., UNSC Resolution 853, 29 July 1993, UN Doc. S/RES/853 (1993). For more relevant 
UNSC resolutions see International Humanitarian Law Databases, Customary IHL, Rule 55, 
<https://ihl-databases.icrc.org/en/customary-ihl/v1/rule55#refFn_E763511D_00020> (accessed 
30 April 2023).
84 See, e.g., UNSC Resolution 688, 5 April 1991, UN Doc. S/RES/688, § 3.
85 UNSC Resolution 2216 (2015), UN Doc. S/RES/2216 (2015), 14 April 2015.
86 UNGA Resolution 46/182, 19 December 1991, UN Doc. 46/182, Annex, Guiding Principles, § 6.
87 See below, section 5.3.
232
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review, voL. LXXXiv, 2024
Views on such situations are diverse. Some scholars contend that the host State in such 
situations is not concerned with humanitarian relief operations provided in opposition 
areas, especially when a non-state actor controls part of the territory of a State and takes 
over governance functions.88 The prevailing view, however, seems to be that the strategic 
consent “of the High Contracting Party concerned”89 as framed in Protocol II refers to 
the State party to the conflict.90 Deriving from the principle of sovereignty over a State’s 
territory, it is, therefore, the prerogative of a State to give consent, even on territories that 
it does not fully control. Although this is highly controversial in situations where the 
civilian population suffers from the lack of necessities, in practice, humanitarian organi-
sations will typically require the consent of the concerned State to provide humanitarian 
assistance on territories controlled by non-state actors.91
Based on the foregoing, the notion of strategic consent seems to be at the heart of 
the international humanitarian law regime governing humanitarian assistance in times 
of armed conflicts. However, as will be explained below, at the operational level, parties 
to the conflict may, for valid reasons (e.g., military necessity), deny operational consent 
to the provision of humanitarian assistance in a particular situation.92
It has been explained in this section that the notion of consent lies at the heart of the 
humanitarian assistance regimes. As a ‘hallmark’ of these regimes, lack of consent is often 
the major practical limitation to humanitarian relief operations.93 However, as is well 
known, the principle of sovereignty, from which the notion of consent derives, cannot 
be perceived as unlimited.94 In the context of humanitarian assistance, this translates into 
the prohibition of arbitrarily withholding consent. In the following two sections, this 
paper will distinguish between occasions of legally justified withholding of operational 
consent on the one hand and the prohibition of arbitrarily withholding strategic consent 
to humanitarian assistance on the other. While States, as parties to the conflict, may 
rely on military necessity to withhold operational consent to humanitarian assistance, 
it will be explained that this is not possible at the strategic level. The final section will 
explain that while arbitrarily withholding consent at the strategic level typically amounts 
to a violation of international law, such withholding of consent cannot be considered a 
legal justification for a per se argument on the legality of non-consensual humanitarian 
assistance. Rather, the underlying violation (arbitrarily withholding consent) triggers the 
88 Bothe, 1982, p. 696; Barber, 2009, pp. 384–385.
89 Protocol II, Article 18(2).
90 See, e.g., Akande in Gillard, 2016, p. 17; Gal, 2017, p. 45.
91 This question was especially pertinent in the context of Syria. See Landgren et al., 2023.
92 Similarly, Ryngaert, 2013, pp. 6–9.
93 Ryngaert, 2013, p. 9.
94 Discussions on “sovereignty as responsibility” or “relative sovereignty” were especially brought for-
ward in the context of the principle of the Responsibility to Protect. See Sancin, 2010, pp. 33–49.
233
Maruša T. Veber – Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
application of relevant secondary rules of international law, under which non-consensual 
provision of humanitarian assistance could be justified.
5. Withholding of the Operational and Strategic Consent to 
Humanitarian Assistance
When analysing the possible withholding of consent to humanitarian assistance, one 
has to distinguish between the following two situations:
1. Withholding of operational consent under international humanitarian law, whereby 
parties to a conflict may have valid legal reasons for withholding consent because 
humanitarian assistance, for example, does not comply with their technical modalities 
of humanitarian assistance (which is embedded in most relevant primary rules on 
humanitarian assistance) or due to military necessity; and
2. The possible arbitrary withholding of strategic consent, which is prohibited under the 
analysed humanitarian assistance regimes.
5.1. Withholding of Operational Consent
It is generally acknowledged that States and parties to conflicts may withhold their 
operational consent to humanitarian assistance under certain circumstances. In this 
respect, consent to humanitarian assistance may be lawfully withheld in cases where 
humanitarian assistance is not aligned with the prescribed technical modalities estab-
lished by the aid-receiving State. Consent may also be withheld for imperative reasons 
of security if, for example, foreign relief personnel could hamper military operations in 
the concerned State,95 as well as on occasion of non-compliance with the principles of 
humanity, neutrality, impartiality,96 and non-discrimination by the external aid provider.
For example, under international humanitarian law, parties to the conflict may exer-
cise control over the relief action.97 In this respect, the aid-receiving State generally has 
the right to “prescribe the technical arrangements” of the humanitarian assistance and 
may make its permission “conditional on the distribution of this assistance being made 
under the local supervision.”98 Moreover, those providing humanitarian relief must not 
“exceed the terms of their mission” and “shall take account of the security requirements 
of the Party in whose territory they are carrying out their duties.”99
95 Akande & Gillard, 2016, p. 499.
96 Stoffels, 2004, pp. 539–544.
97 See, e.g., Fourth Geneva Convention, 1949, Article 23; Protocol I, Article 70(3); Henckaerts, 
2005, Rule 55.
98 Fourth Geneva Convention, Article 23; Protocol I, Article 70(3).
99 Protocol I, Article 71(4).
234
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review, voL. LXXXiv, 2024
Similarly, the ILC explicitly recognised in its work on the protection of persons in the 
event of disasters that “the affected State may place conditions on the provision of external 
assistance.”100 This is also confirmed in practice, whereby international organisations pro-
viding humanitarian assistance to affected States typically sign a specific agreement and ne-
gotiate these technical modalities of humanitarian assistance before they engage with a con-
cerned State. In this sense, they also secure the advance consent of the concerned State.101
Determining the modalities of the provision of humanitarian assistance—i.e. wheth-
er it will include the use of AI—could arguably fall under this ‘technical’ category of 
valid legal reasons to deny humanitarian assistance. In situations where humanitarian 
assistance is rejected due to its AI component, as was the case in Yemen, the question of 
alternative means to provide humanitarian assistance arises. For example, could a UN 
agency assist without relying on AI technology? Arguably, denying the use of AI on the 
territory of a State cannot be seen as breaching rules on humanitarian assistance in in-
stances where adequate alternative solutions exist.
In this respect, the possibility of denying humanitarian assistance due to technical 
modalities, such as the use of AI, seems to be embedded in the primary rules governing 
humanitarian assistance and cannot be seen as a violation of these rules. It has to be 
acknowledged, however, that technical arrangements have to be applied in good faith, 
whereby their imposition or effect must not be arbitrary.102 According to the ICRC, mil-
itary necessity can only “be invoked in exceptional circumstances in order to regulate 
—but not prohibit— humanitarian access, and can only temporarily and geographically 
restrict the freedom of movement of humanitarian personnel.”103 Unjustified withholding 
of operational consent would amount to a violation of relevant rules of international hu-
manitarian law. However, as will be explained below, in instances where such denial would 
be all-encompassing and unjustified and would amount to serious violations of the State’s 
other international obligations relating to the civilian population, and where the use of AI 
would be the only way to distribute such assistance or would be proportionally the most 
appropriate and efficient way to distribute the assistance, one could discuss the issue of the 
arbitrary withholding of consent and the subsequent responsibility of the concerned State.
5.2. Arbitrary Withholding of Strategic Consent
Withdrawal of operational consent due to technical modalities of humanitarian as-
sistance has to be distinguished from the arbitrary withholding of strategic consent to 
100 Draft articles on the protection of persons in the event of disasters, 2016, draft Article 14.
101 See, e.g., Article XI (Assistance agreements), General Regulations and General Rules, WFP (2022) 
<https://executiveboard.wfp.org/document_download/WFP-0000141150> (accessed 20 April 2023).
102 Oxford Guidance on the Law Relating to Humanitarian Relief Operations in Situations of Armed 
Conflict, 2016, § 71.
103 ICRC, 2014, p. 364.
235
Maruša T. Veber – Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
humanitarian assistance. While the first is a right of States deriving from primary rules 
governing humanitarian assistance, strategic consent is a precondition for the delivery of 
humanitarian aid in the first place, and its arbitrary withholding is generally prohibited 
by these same primary rules.104 It seems to be generally accepted that strategic consent to 
humanitarian assistance cannot be arbitrarily withheld.105
The modalities of arbitrariness are, however, not generally defined under internation-
al law. Arbitrariness is, therefore, typically dependent on the circumstances of a concrete 
situation. Some guidance as to arbitrariness can be found under international human 
rights law, whereby the question of whether withholding of consent is necessary and pro-
portionate to achieving legitimate ends is crucial.106 In this respect, arbitrariness has been 
understood as refusing consent in a manner that is “unjustified” and not “in pursuit of 
[a] legitimate aim”107; “unreasonable, unjust, lacking in predictability or […] otherwise 
inappropriate”108 or not pursued for “reasons that are valid and compelling.”109 It is often 
argued that conduct which would violate other obligations of a State under international 
law should be regarded as arbitrary.110 According to Sivakumaran, refusal is arbitrary if it
“results in the violation by a state of its obligations under international law con-
cerning the civilian population in question (such as its human rights obligations), 
or if it violates the principle of necessity and proportionality, or if it discriminates 
against a particular group.”111
In this sense, there seems to be agreement among scholars and institutions that if the 
withholding of consent results in mass atrocities, such as war crimes or crimes against 
humanity, it could arguably be considered arbitrary.112 In this respect, under international 
humanitarian law, a denial of humanitarian assistance to cause, contribute, or perpetuate 
starvation would amount to a violation of the prohibition of starvation of the civilian pop-
104 International Humanitarian Law Databases, Customary IHL, Rule 55, <https://ihl-databases.icrc.
org/en/customary-ihl/v1/rule55#refFn_E763511D_00020> (accessed 30 April 2023); Draft articles 
on the protection of persons in the event of disasters, 2016, draft Article 13; Institute of International 
Law, Santiago de Compostela Resolution, 1989, Article 5; see also IIL Resolution, 2003.
105 Report of the Secretary-General on the protection of civilians in armed conflict, UN Doc. S 
/2013/689, 22 November 2013, § 58; Oxford Guidance on the Law Relating to Humanitarian 
Relief Operations in Situations of Armed Conflict, 2016, section E; Akande and Gillard, 2016, pp. 
489 ff; Henckaerts & Doswald-Beck, 2005, p. 197; IIL Resolution, 2003, § VIII; Draft articles on 
the protection of persons in the event of disasters, 2016, draft Article 13(2).
106 See also Akande & Gillard, 2016, pp. 498–499 and 505–507.
107 Sivakumaran, 2015, pp. 517–521.
108 Akande & Gillard, 2016, p. 22.
109 Gillard, 2013, p. 360.
110 Akande & Gilliard, 2016, pp. 494–495.
111 Sivakumaran, 2015, p. 521.
112 Rottensteiner, 1999, pp. 555–581; IIL Resolution, 2003, § VIII.
236
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review, voL. LXXXiv, 2024
ulation as a method of warfare113 and may also amount to a war crime under international 
criminal law.114 Moreover, “intentional inflictions of conditions of life, inter alia the dep-
rivation of access to food and medicine, calculated to bring about the destruction of part 
of a population” may amount to extermination as a crime against humanity.115 Similarly, 
systematic rejection of humanitarian assistance in areas populated by a particular ethnic 
group would amount to a violation of the rule prohibiting adverse distinction under in-
ternational humanitarian law116 and the prohibition of discrimination under international 
human rights law,117 and could possibly amount to a crime against humanity.118
The question of the withholding of consent to humanitarian assistance has been ex-
tensively addressed outside the AI context, especially in cases where it resulted in gross vi-
olations of international humanitarian law and international human rights law. This was 
the case in Ethiopia, where the Mengistu regime banned the movement of relief supplies 
during the famine that emerged in 1989119 and more recently, in the case of Syria’s denial 
of consent to international humanitarian aid on the territories controlled by non-state 
actors.120 In these situations, it has been argued that arbitrary denial of humanitarian 
assistance violates the aforementioned rules of international humanitarian law and inter-
national human rights law governing humanitarian assistance.121 In the past, the Human 
Rights Committee has considered arbitrary denial of humanitarian assistance as violating 
international human rights obligations of States, including the right to life.122
113 Protocol I, Article 54(1); Protocol II, Article 14; Akande & Gilliard, 2016, pp. 495–496.
114 Rome Statute of the International Criminal Court, UNTS 2187, 17 July 1998, EIF 1 July 2002, p. 
3 (Rome Statute), Article 8(2)(b)(xxv).
115 Rome Statute, Articles 7(1)(b) and 7(2)(b); ILC Draft articles on Prevention and Punishment of 
Crimes Against Humanity, with commentaries, YILC 2019, vol. II, Part Two, p. 28. See also na-
tional criminal legislations, e.g., Criminal Code of the Republic of Slovenia, Official Gazette of the 
Republic of Slovenia, No. 50/12 – official consolidated version, 6/16 – corr., 54/15, 38/16, 27/17, 
23/20, 91/20, 95/21, 186/21, 105/22 – ZZNŠPP and 16/23, Article 101.
116 Common Article 3 of the Geneva Conventions. See also Geneva Convention Relative to the 
Treatment of Prisoners of War (Third Geneva Convention), 12 August 1949, 75 UNTS 135, Article 
16; Fourth Geneva Convention, Article 13; Protocol I, Article 75(1); Protocol II, Article 4(2).
117 See, e.g., ICCPR, Article 26; Akande & Gilliard, 2016, p. 497.
118 Rome Statute, Article 7(1)(h).
119 International Humanitarian Law Databases, Customary IHL, Rule 55, <https://ihl-databases.icrc.
org/en/customary-ihl/v1/rule55#refFn_E763511D_00020> (accessed 30 April 2023).
120 Ryngaert, 2013, pp. 5–19. See below, section 5.3.
121 UNSC, Statement by the President of the Security Council, UN Doc. S/PRST/2013/15, 2 October 
2013. See also UNSC Resolution 2216 (2015), UN Doc. S/RES/2216 (2015), 14 April 2015; 
UNGA Resolution 68/182, UN Doc. A/RES/68/182, 30 January 2014, § 14.
122 See, e.g., Human Rights Committee, Concluding observations on the fourth periodic report of the 
Sudan, UN Doc. CCPR/C/SDN/CO/4, 19 August 2014, § 8.
237
Maruša T. Veber – Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
As already explained, conditioning consent to humanitarian assistance with certain 
technical reasons, i.e. the non-use of a particular AI system, seems to be supported by the 
relevant primary rules of international law governing humanitarian assistance. However, 
when such denial would amount to serious violations of the State’s other international 
obligations relating to the civilian population, and where the use of AI would be the only 
way to distribute such assistance or would be proportionally the most appropriate and 
efficient way to distribute the assistance, the issue of the arbitrariness of such withhold-
ing of consent comes to the forefront. The assessment of arbitrariness has to be made 
on a case-by-case basis, taking into consideration the above-mentioned elements and 
is fraught with difficulty in the decentralised international legal reality. While on these 
occasions the arbitrariness of the withholding of consent seems self-evident, stemming 
from the gravity of the underlying breach, this is not always the case. Under international 
human rights law, there may be a need for a more nuanced balance between the provi-
sion of humanitarian assistance and the realisation of the right to life and the right to 
food on the one hand and, for example, the right to privacy, which the use of AI on the 
territory of the State concerned may undermine, on the other. It is beyond the scope of 
this paper to provide a detailed overview of the cases and arguments relating to the arbi-
trary withdrawal of consent. For our discussion, it is important to determine what legal 
consequences stem from the arbitrary withdrawal of consent to humanitarian assistance: 
does it result in the per se legality of non-consensual humanitarian aid, or does it trigger 
justification of non-consensual provision of humanitarian aid under the secondary rules 
of international law? This is important because it essentially determines the legal analysis 
and course of action to be undertaken in such situations by international organisations.
5.3. Legal Consequences of Arbitrary Withholding the Strategic Consent
Some argue that in instances of arbitrary withholding of consent, the provision of 
non-consensual humanitarian assistance is to be considered as per se lawful.123 These 
scholars seem to make the argument that there exists, at the level of the primary rules 
governing humanitarian assistance, a customary international legal rule allowing for the 
non-consensual provision of humanitarian relief in such cases. This view became par-
ticularly vocal in the case of Syria, whereby, despite the deterioration of the humanitar-
ian situation due to an ongoing conflict in the country, the Syrian government refused 
to give consent to “cross-border” operations, to reach more than three million people 
who were located in remote areas. Already in 2014, a coalition of international lawyers 
made a statement arguing that “there is no legal barrier to the UN directly undertaking 
123 Barber, 2023, p. 1; Barber, 2009, pp. 371–397; Stoffels, 2004, p. 536; American Relief Coalition 
for Syria, 2022, pp. 25 ff.
238
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review, voL. LXXXiv, 2024
cross-border humanitarian operations”124 to opposition-controlled areas because they 
meet all conditions for legality, neutrality, impartiality, and non-discrimination, and due 
to the prior arbitrary withholding of consent by Syria causing a serious humanitarian 
situation in the country.125
This view was reinstated in 2023, after a devastating earthquake in southern Turkey, 
which seriously affected thousands of people in northwest Syria, destroying the border 
crossing between the countries, Babal-Hawa, the only crossing the UN Security Council 
(UNSC) has authorised for humanitarian assistance to the opposition-held territory 
in Syria. This caused significant delays in international aid deliveries. Some scholars126 
and a group of eminent academics and professionals adopted the statement “There is 
Still No Legal Barrier to UN Cross-Border Operations in Syria Without a UN Security 
Council Mandate,”127 in which they argue, among other things, that refusal to permit 
cross-border aid in this situation is unlawful as it is arbitrary and necessitates continuous 
cross-border provision of aid to prevent possible distress, strife, and starvation.128
Without a doubt, in such situations, where the lives of millions of people who rely 
on cross-border aid are put at risk, allowing the non-consensual delivery of (possibly 
AI-supported) humanitarian aid seems reasonable and humane. However, it is argued 
here that legal analysis in these cases should nevertheless be nuanced. Rather than argu-
ing that non-consensual humanitarian assistance is per se lawful in such situations, one 
should carefully analyse and apply relevant primary and secondary rules of international 
law accordingly. In particular, the question that has to be addressed in such situations is 
whether a non-consensual, “clandestine”, AI-supported humanitarian assistance opera-
tion would be in line with the principles of sovereignty and territorial integrity and the 
principle of the prohibition of intervention in the internal affairs of a State. And subse-
quently, whether a violation of these primary rules could be justified under international 
law on the basis of the secondary rules of international law, such as countermeasures.
6. Non-Consensual AI-Supported Humanitarian Aid and the Principles 
of Non-Intervention and Sovereignty
The provision of non-consensual humanitarian assistance is in fundamental tension 
with two fundamental principles of international law: the principle of sovereignty and its 
corollary, the principle of non-intervention. This conclusion is embedded in the primary 
124 The Guardian, 2014.
125 Ibid.
126 Barber, 2023, p. 1.
127 There is Still No Legal Barrier to UN Cross-Border Operations in Syria Without a UN Security 
Council Mandat, 2023.
128 Ibid. See also Barber, 2023; Sproson & Olabi, 2023.
239
Maruša T. Veber – Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
rules governing humanitarian assistance, which, as explained above, are preconditioned 
on the consent of the concerned State.129
Discussions on the provision of humanitarian aid are generally centred on the princi-
ple of non-intervention as interpreted by the International Court of Justice (ICJ) in the 
Nicaragua case. This principle, often labelled as controversial,130 is commonly perceived as 
consisting of two elements: (1) the act in question relates to the internal or external affairs 
of the targeted State, and (2) the act is coercive in nature.131 In the Nicaragua case, the 
Court analysed these two elements when discussing the support of the US to the Contras 
in the form of financial support, training, supply of weapons, intelligence, and logistic 
support. The ICJ considered this support a clear breach of the principle of non-interven-
tion due to its purpose, i.e. coercing Nicaragua and supporting the Contras to overthrow 
the government. In contrast, however, the Court stressed that the provision of human-
itarian assistance cannot be considered as violating the principle of non-intervention:
“The Court has however taken note that, with effect from the beginning of the 
United States governmental financial year 1985, namely 1 October 1984, the 
United States Congress has restricted the use of the funds appropriated for assistan-
ce to the contras to ‘humanitarian assistance’ (paragraph 97 above). There can be no 
doubt that the provision of strictly humanitarian aid to persons or forces in another 
country, whatever their political affiliations or objectives, cannot be regarded as 
unlawful intervention, or as in any other way contrary to international law.”132
The ICJ further stressed that there cannot be an intervention in internal affairs in cas-
es where humanitarian assistance is limited to preventing and alleviating human suffer-
ing, protecting life and health, and ensuring respect for human beings, whereby it must 
be given without discrimination.133 This passage is commonly cited as confirming that 
the provision of humanitarian assistance cannot be considered as violating the principle 
of non-intervention.134 However, it is argued here, that the ICJ’s statement necessitates 
a more nuanced analysis.
The question that this passage raises is whether the ICJ was referring to the provision of 
humanitarian assistance without crossing the border of the concerned State (Nicaragua) 
or to humanitarian assistance, including direct engagement with relief operations inside 
the country. Given the context in which the Court reached its conclusions—where it had 
previously found that US aid and support to the Contras (without crossing the border) 
was contrary to the principle of non-intervention due to the purposes of such aid, i.e. 
129 Gillard, 2013, p. 369; Stoffels, 2004, p. 535.
130 Jamnejad & Wood, 2009, p. 346.
131 Schmitt (ed.), 2017, p. 314.
132 Case Concerning Military and Paramilitary Activities in and Against Nicaragua (Nicaragua v. United 
States of America), Judgment (1986), ICJ Rep. 14, § 242 (Nicaragua case).
133 Ibid., § 243.
134 Barber, 2020, pp. 9–10.
240
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review, voL. LXXXiv, 2024
coercing Nicaragua in respect of matters in which each State is permitted by the principle 
of state sovereignty to decide freely, and the purpose of Contras to overthrow the gov-
ernment of Nicaragua—one could deduce that the Court was referring to humanitarian 
assistance without crossing the border. Thus, at the centre of the decision was not the 
issue of consent and physical intrusion into the territory, but rather the purpose and 
nature of the aid provided by the US,135 without crossing the border.
An essential element of the principle of non-intervention is coercion, whereby a “pro-
hibited intervention must constitute an attempt to coerce the targeted State by directly 
or indirectly interfering in the internal or external affairs of this State.”136 In this respect, 
the element of ‘coercion’ in the principle of non-intervention has two facets: (1) coercion 
in a physical, direct, sense, e.g. with the use of force, and (2) non-physical, indirect sense, 
as was the case with, e.g., the provision of financial support.137
In cases of non-consensual humanitarian assistance, neither of these elements will 
typically be met, as the mere provision of impartial humanitarian assistance does not by 
itself constitute a coercive act aimed at affecting the freedom of decision of the targeted 
State.138 The Nicaragua decision, which asserts that humanitarian aid does not violate the 
principle of non-intervention because it does not seek to coerce the free will of the State 
concerned, must, therefore, be read in this context.
Against this background, one could hardly conclude that the cited passage of the 
ICJ could be understood as permitting the cross-border provision of humanitarian relief 
inside the affected State without the consent of the concerned State under international 
law.139 Indeed, an offer of humanitarian assistance cannot be considered a violation of the 
principle of non-intervention, as it does not fulfil the “coerciveness” criterion. However, 
for it to be physically provided on the territory of a concerned State, that State’s consent 
is necessary. This is because such physical non-consensual provision of humanitarian as-
sistance would breach another principle of international law: the principle of sovereignty.
Non-consensual, “mere” physical intrusion into the territory of another State, with-
out a coercive element and the aim of affecting the internal or external affairs of the 
concerned State, is typically analysed in the context of the principle of sovereignty, which 
is a separate principle, albeit intrinsically related to the principle of non-intervention.140 
According to this central141 principle of international law, States have supreme author-
ity over their land, territory and appurtenances (e.g., internal waters, territorial seas, 
135 Nicaragua case, 1986, §§ 239–244.
136 Delerue, 2020, p. 235.
137 Nicaragua case, 1986, § 205.
138 Similarly Delerue in relation to cyber espionage and the principle of non-intervention. Delerue, 
2020, p. 258.
139 For a similar conclusion, see: Gillard, 2013, p. 370; Sproson & Olabi, 2023.
140 Nicaragua case, 1986, § 202.
141 Ibid., § 263.
241
Maruša T. Veber – Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
archipelagic waters, airspace, and subsoil).142 It is in the context of this principle that 
clandestine, non-consensual operations, while being physically present in the territory 
of a foreign State, are typically discussed.143 This can also be buttressed by the fact that 
in the Nicaragua case, violations of sovereignty by way of physical intrusion—e.g. by 
unauthorised overflights of Nicaragua’s territory by aircraft belonging to or under the 
control of the government of another State (the US)—were addressed separately from 
the principle of non-intervention.144
In a similar way conducting espionage by State agents on the territory of another 
State violates the principle of sovereignty145 (and on some occasions, if the criteria are 
fulfilled, also the principle of non-intervention146), the non-consensual provision of AI-
supported humanitarian assistance in the territory of the concerned State also amounts 
to a violation of the principle of sovereignty. According to Buchan, any “non-consensual 
incursion by one State into the territory of another State violates the rule of territorial 
sovereignty, regardless of whether that infraction produces damage.”147 Similarly, Deleure 
concludes that “an unauthorised act by a State on the territory of the targeted State 
violates the territorial sovereignty of the latter”148 and that damage need not occur.149 
Therefore, unauthorised physical intrusions, even of a neutral and humanitarian nature, 
violate the principle of sovereignty. It also has to be acknowledged, however, that wheth-
er the provision of humanitarian aid in a concrete case violates the principle of sovereign-
ty and non-intervention must be assessed on a case-by-case basis.150
Finally, it has to be acknowledged that the non-consensual provision of humanitarian 
assistance with the use of AI also triggers another facet of the principles of non-inter-
vention and sovereignty: the question of possible unauthorised, clandestine gathering of 
large amounts of data on the territory of a concerned State through AI systems. In the 
past, such clandestine activities were considered to amount to a violation of the principle 
of sovereignty and non-intervention by States and courts. For example, in 2008, the 
Federal Court of Canada published its response to a request from the Canadian Security 
Intelligence Service (CSIS) to approve a warrant under Section 12 of the Canadian 
Security Intelligence Service Act 1984 to conduct surveillance against individuals locat-
142 Ibid., § 212; Delerue, 2020, p. 200.
143 Jennings & Watts, 2008, pp. 385–386.
144 See Nicaragua case, 1986, §§ 251 ff.
145 Delerue, 2020, p. 212.
146 Wright, 1962.
147 Buchan, 2021, p. 51.
148 Delerue, 2020, p. 212.
149 Ibid., pp. 213 and 215–219.
150 It has been argued, for example, with respect to the situation in Syria that UN agencies are not 
providing cross-border aid. Sproson & Olabi, 2023, p. 4.
242
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review, voL. LXXXiv, 2024
ed within the territory of other States. Under Canadian law, the Court could only issue 
the warrant if the activities being authorised were compliant with international law. In 
refusing to grant the warrant, the Court observed that:
“The intrusive activities […] are activities that clearly impinge upon the above-
-stated principles of territorial sovereign equality and non-intervention […] By 
authorizing such activities, the warrant would therefore be authorizing activities 
that are inconsistent with and likely to breach the binding customary principles of 
territorial sovereign equality and non-intervention, by the comity of nations. These 
prohibitive rules of customary international law […] have evolved to protect the 
sovereignty of nation states against interference from other states.”151
In the context of cyber espionage, other States like Argentina, Bolivia, Brazil, Uruguay, 
and Venezuela condemned the clandestine activities of the US as unacceptable behaviour 
that violates their sovereignty. Against this background, scholars have concluded that 
“pulling” data without the consent of a concerned State is contrary to international law, 
in particular the principle of sovereignty.152
To conclude, the non-consensual provision of humanitarian assistance is potentially 
in violation of two fundamental principles of international law: the principle of sover-
eignty and the principle of non-intervention. However, as will be explained in the last 
part of this paper, the consent of the concerned State could be substituted by a UNSC 
authorisation under Chapter VII of the UN Charter or justified under the secondary 
rules of international law, countermeasures in particular.
7. Possible Legal Justifications for the Non-consensual Provision of 
Humanitarian Aid Under International Law
Indeed, while the arbitrary withholding of consent could be considered as violating 
the rules governing humanitarian assistance, in particular international humanitarian 
law and international human rights law, this unlawfulness itself, does not justify the 
non-consensual delivery of humanitarian assistance at the level of the primary rules due 
to its tensions with fundamental international legal principles. Rather, recourse to the 
UNSC or to secondary rules of international law is necessary to justify non-consensual 
humanitarian assistance in cases of withholding of consent.
7.1. The Authorisation of the UNSC
In situations where humanitarian assistance is being arbitrarily withheld, deliberately 
denied or obstructed, and where such denial may constitute a threat to international 
151 Federal Court, Canadian security intelligence service act (re) (f.c.), SCRS-10-07, 2008 CF 301.
152 Buchan, 2021, pp. 54–55.
243
Maruša T. Veber – Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
peace and security, the UNSC may adopt appropriate measures under Chapter VII of 
the UN Charter153 to remedy the situation. The role of the UNSC in substituting the 
consent of the State for the provision of humanitarian assistance has been extensively 
analysed in the context of the situation in Syria, whereby the Syrian government denied 
humanitarian access to civilians in opposition-controlled areas. In response to this, the 
UNSC first demanded the Syrian authorities to allow the delivery of humanitarian as-
sistance.154 Failing to do so, and due to the deterioration of the humanitarian situation 
in Syria, whereby the number of people in need of assistance exceeded 10 million, and 
disturbed by the “continued, arbitrary and unjustified withholding of consent to re-
lief operations,” the UNSC with resolution 2165 (2014),155 authorised UN agencies to 
provide humanitarian assistance in Syria through four designated international border 
crossings without the consent of the Syrian government.156 This resolution was continu-
ously renewed until 2019, whereas in 2020, due to the veto of Russia and China, which 
were concerned over the sovereignty of Syria,157 the authorisation for the cross-border 
humanitarian operation was reduced to only include one international border crossing 
(Turkey).158 This authorisation was considered necessary since the Syrian government 
restricted the delivery of humanitarian assistance to the areas not under its control.159
The UNSC authorisation to substitute the consent of a concerned State is a reasona-
ble solution in cases of arbitrary withholding of consent by a concerned State. However, 
in cases where the modalities and technical details for the provision of humanitarian 
assistance are in question and the primary reason for denying AI-supported humanitar-
ian assistance, it seems less likely that the UNSC would intervene. Only in a situation 
where there is no appropriate technical alternative to the use of AI, and the consequences 
of refusing to allow the use of AI when distributing humanitarian assistance cause sig-
nificant damage to the civilian population in need—amounting, for example, to star-
vation—would the adoption of a UNSC resolution be reasonable. Rather, the issue of 
technical modalities should be negotiated among concerned parties, i.e. the humanitar-
ian organisation and the State. It is only reasonable that the humanitarian organisation 
tries to comply with the technical requirements proposed by the host State to distribute 
assistance to those in need.
153 UNSC Resolution 1265, 17 September 1999, UN Doc. S/RES/1265; IIL Resolution 2003, § VII(3).
154 UNSC Resolution 2139, 22 February 2014, UN Doc. S/RES/2139, 2014. See also Presidential 
Statement of 2 October 2013, UN Doc. S/PRST/2013/15.
155 UNSC Resolution 2165, 14 July 2014, UN Doc. S/RES/2165, 2014.
156 UNSC Resolution 2165, § 2.
157 UNSC SC/14066, 20 December 2019, UN Doc. SC/14066.
158 See UNSC Resolution 2504, 10 January 2020, UN Doc. S/RES/2504; UNSC Resolution 2533,13 
July 2020, UN Doc. S/RES/2533.
159 Barber, 2020, p. 1
244
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review, voL. LXXXiv, 2024
7.2. Necessity
Under secondary rules of international law, the most commonly proposed legal jus-
tifications for the provision of non-consensual humanitarian assistance are necessity and 
countermeasures, as one of the circumstances precluding wrongfulness of conduct that 
would otherwise not conform with the international obligations of the concerned hu-
manitarian organisation. In other words, because the non-consensual provision of hu-
manitarian assistance arguably violates the principle of sovereignty, this last part of the 
paper addresses the question of whether international humanitarian organisations could 
justify such a breach by relying on necessity or countermeasures as circumstances pre-
cluding wrongfulness.
According to Article 25 ARIO, an international organisation may invoke necessity as 
a ground for precluding the wrongfulness of an act not in conformity with an interna-
tional obligation of that organisation, where that act is
“the only means for the organization to safeguard against a grave and imminent 
peril an essential interest of its member States or the international community as a 
whole, when the organization has, in accordance with international law, the fun-
ction to protect that interest.”160
Additionally, to be able to rely on necessity, an international organisation must not 
seriously impair an essential interest of the concerned State.
The idea of the ILC, when introducing this circumstance, was that it would be used 
in exceptional and limited cases, under narrowly defined conditions161, where an irrec-
oncilable conflict between an essential interest on the one hand and the obligation of the 
concerned State or international organisation invoking necessity on the other exists.162 
The ILC, however, explicitly stated in the commentaries that (forcible) humanitarian in-
tervention cannot be justified on necessity as a circumstance precluding wrongfulness.163 
What is more, the ILC concluded that the plea of necessity should not be invocable 
by international organisations as widely as by States and thereby limited the possibility 
of international organisations to rely on the necessity to instances where the essential 
interest of its member States or the international community as a whole is at stake, and 
the organisation has, in accordance with international law, the function to protect that 
interest.164 Against this background, three conditions have to be met for an international 
organization to invoke necessity:
160 Articles on Responsibility of International Organizations (ARIO), YILC 2011, vol. II, Part Two, 
Article 25(1)(a).
161 Commentary to Article 25 ARIO, § 1.
162 Commentary to Article 25, Articles on Responsibility of States for Internationally Wrongful Acts 
(ARSIWA), YILC, 2001, vol. II (Part Two), p. 80, §§ 1–2.
163 Commentary to Article 25 ARSIWA, § 21.
164 Commentary to Article 25 ARIO, p. 52, § 4.
245
Maruša T. Veber – Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
1. Grave and imminent peril to an essential interest of its member States or the interna-
tional community as a whole;
2. A concerned international organisation has the function to protect that interest;
3. The course of action, i.e. non-consensual humanitarian assistance, is the only available 
way to safeguard that interest.
There is, however, very little practice whereby international organisations would rely 
on the notion of necessity to justify their actions.165
Regarding the first criterion, a situation whereby denial of humanitarian assistance 
leads to suffering and grave violations of the rights of the civilian population, possi-
bly amounting to war crimes or crimes against humanity, could be considered as erga 
omnes obligations, and thus in the essential interest of the international community as 
a whole.166 Regarding the second condition, UN agencies generally are mandated with 
the provision of humanitarian assistance and have the function to protect the concerned 
essential interest (helping the civilian population in need, preventing humanitarian ca-
tastrophes167). While necessity has been considered by scholars as a possible legal basis 
for the provision of non-consensual humanitarian assistance,168 its application to AI-
supported humanitarian assistance is questionable. In particular, it is the third condition 
that is problematic in our context, as typically, an alternative to AI-supported human-
itarian assistance should be available to international organisations. If there are other 
(non-AI) means by which humanitarian assistance could be delivered, even if they are 
more costly or less convenient, this condition will not be met.169
7.3. Countermeasures
Whether non-consensual humanitarian relief operations can be characterised as counter-
measures has already been considered (albeit briefly) by other scholars.170 Countermeasures 
are one of the circumstances precluding wrongfulness, allowing for the response to a previ-
ous breach of international law with the adoption of measures that would otherwise them-
165 Commentary to Article 25 ARIO, p. 51, § 2.
166 For discussions whether ensuring safety of civilian population and severe suffering of civilian pop-
ulation amounts to the essential interest, see: Commentary to Article 25 ARSIWA, p. 83; Barber, 
2023, p. 3; Gillard, 2013, p. 373; Ryngaert, 2013, p. 15.
167 See, e.g., Article II (The purposes and functions of WFP), General Regulations and General Rules, 
WFP, 2022, <https://executiveboard.wfp.org/document_download/WFP-0000141150> (accessed 
20 April 2023).
168 Barber, 2023, p. 3; American Relief Coalition for Syria, 2022, pp. 37–43.
169 Commentary to Article 25 ARSIWA, p. 83, § 15.
170 See, e.g, Akande & Gillard, 2016, pp. 54–55; Ryngaert, 2013, p. 15; Stoffels, 2004, p. 537.
246
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review, voL. LXXXiv, 2024
selves be contrary to international law.171 Countermeasures aim to ensure the cessation of 
the alleged breach and, where appropriate, ensure reparation for injury.172 They are thus of 
temporary or provisional character, aiming at the restoration of legality. If the responsible 
subject complies with its obligations of cessation and reparation, the countermeasures are 
to be discontinued, and the performance of the obligation resumed.173
Countermeasures may be adopted by international organisations to protect their in-
dividual interest when they are injured by the previous internationally wrongful act,174 
or to safeguard a general interest of a sort.175 It is the latter, more controversial type of 
countermeasures that could be relevant for our discussion. In brief, the idea of the so-
called third-party countermeasures is that it gives States and international organisations 
the entitlement to invoke responsibility and adopt countermeasures in instances where 
they are not directly injured by a prior breach of international law, that is, in response to 
erga omnes (partes) obligations. Due to their fundamental character these obligations are 
“the concern of all States”, because “all States can be held to have a legal interest in their 
protection”;176 they are owed to the international community as a whole. The question 
of the legality of the adoption of countermeasures in response to violations of erga omnes 
obligations as codified by the ILC, has received a lot of attention among scholars.177 It is 
not the purpose of the present research to further explore extensive debates on the legality 
of the adoption of countermeasures in response to violations of erga omnes obligations. It 
is important to note that from 2001 onwards acceptance of the legality of such measures, 
also against the background of increased practice, is becoming firmly established amongst 
international lawyers178 and other important professional organisations on international 
law.179 It is, therefore, premised here, that such an entitlement exists in international law.
In cases of withholding of consent to humanitarian assistance, it is the civilian pop-
ulation that suffers and is, therefore, directly affected by the activities of their host State. 
While the civilian population as such has limited options if invoking the responsibility of 
171 Article 22 and Part three, Chapter II ARSIWA; Article 22 and Part four, Chapter II ARIO. Naulilaa 
Incident Arbitration (Portugal v. Germany), 31 July 1928, RIAA, vol. 2 (UN publications 1949), 
p. 1012; Gabčíkovo-Nagymaros Project (Hungary v. Slovakia), Judgment (1997) ICJ Rep. 7, pp. 
55–56; Air Services Agreement of 27 March 1946 between the United States of America and France, 9 
December 1978.
172 Articles 22 and 49 ARSIWA; Articles 22 and 51 ARIO.
173 Commentary to Article 49, ARSIWA, pp. 130–131, § 7.
174 Article 22 ARIO (and Article 22 ARSIWA). 42. ARSIWA and Article 43. ARIO.
175 Article 48 and 54 ARSIWA and Article 49 and 57 ARIO.
176 Barcelona Traction, Light and Power Company, Limited, (Belgium v. Spain), Judgment (1970) ICJ 
Rep. 3, §§ 33–34.
177 Frowein, 1987; Alland, 2002; Tams, 2005, pp. 198–251; Gaja, 2011; Dawidowicz, 2017.
178 See T. Veber, 2022, p. 311, footnote 2375.
179 IIL, Resolution on Obligations erga omnes in international law, Krakow, 2005.
247
Maruša T. Veber – Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
this State, in cases where denial of humanitarian assistance amounts to a violation of an 
erga omnes obligation, e.g. crime against humanity, genocide or war crime, it is the other 
actors of the international community that have the entitlement to react on such occa-
sions through the adoption of countermeasures in the form of non-consensual delivery of 
humanitarian assistance. As already explained, non-consensual humanitarian assistance 
generally amounts to a violation of the sovereignty of the concerned State. However, in 
situations where denial of AI-supported humanitarian assistance leads to a violation of 
an erga omnes obligation, States and international organisations are entitled to provide 
non-consensual humanitarian assistance through the adoption of countermeasures.180
8. Concluding Remarks
Careful analysis of relevant regimes governing humanitarian assistance reveals that 
the consent of the concerned State continues to have a central role in the general human-
itarian assistance regime, the international human rights regime and the international 
humanitarian law regime. The notion of consent lies at the heart of these rules, and 
subsequently the lack of consent is often the major practical limitation to humanitarian 
relief operations. This paper distinguished between two different types of consent to 
AI-supported humanitarian assistance: strategic consent and operational consent. The 
former refers to the general consent of a State to the delivery of humanitarian assistance 
on its territory, while the latter refers to the consent required at the operational level for 
the delivery of a particular type of humanitarian assistance in a specific geographically 
defined area. States can have valid legal reasons for withholding the operational con-
sent, including because humanitarian assistance, for example, does not comply with their 
technical modalities of humanitarian assistance, which is embedded in most relevant 
primary rules on humanitarian assistance. Against this background, States may validly 
withhold operational consent to AI-supported humanitarian assistance and request an 
alternative (non-AI) distribution of humanitarian assistance. On the other hand, strate-
gic consent, which is a prerequisite to the delivery of humanitarian aid to a particular 
State, cannot be arbitrarily withheld, which is the case where such denial would amount 
to serious violations of the State’s other international obligations relating to the civilian 
population, and where the use of AI would be the only way to distribute such assistance 
or would be proportionally the most appropriate and efficient way to distribute the as-
sistance. In such an instance, the issue of the responsibility of a concerned State arises, 
and subsequently also possible non-consensual AI-supported humanitarian assistance.
It has been explained that the provision of non-consensual humanitarian aid cannot 
be considered as per se legal under international law. Careful analysis of relevant ICJ case-
180 For a view that the law of countermeasures cannot be applied to non-consensual humanitarian 
assistance, see Stoffels, 2004, p. 536.
248
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review, voL. LXXXiv, 2024
law and State practice reveals that the non-consensual provision of humanitarian assis-
tance would amount to a violation of the principle of sovereignty and non-intervention. 
However, non-consensual humanitarian assistance could nevertheless be justified by a 
UNSC authorisation or under the secondary rules of international law, countermeasures 
in particular. This latter possibility is limited to instances whereby denial of AI-supported 
humanitarian assistance would simultaneously lead to a violation of an erga omnes obli-
gation, thereby triggering the entitlement for the provision of non-consensual humani-
tarian assistance through the adoption of countermeasures.
249
Maruša T. Veber – Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
References
Akande, D., & Gillard, E.-C. (2016) ‘Arbitrary Withholding of Consent to Huma-
nitarian Relief Operations in Armed Conflict’, International Law Studies, Vol. 
92, pp. 483–511.
Alland, D. (2002) ‘Countermeasures of General Interest’, European Journal of Inter-
national Law, Vol. 13, No. 5, pp. 1221–1239.
American Relief Coalition for Syria (2022), 2014 is not 2022: why the continuation of UN-
coordinated cross-border aid into Syria absent a UN Security Council resolution 
is lawful, <https://www.crossborderislegal.org/_files/ugd/166212_22269a6fb 
394455f89a2f25d568c2819.pdf> (accessed 20 April 2023).
Barber, R. (2009) ‘Facilitating humanitarian assistance in international humanitarian 
and human rights law’, International Review of the Red Cross, Vol. 91, No. 874, 
pp. 371–397.
Barber, R. (2020) ‘Does International Law Permit the Provision of Humanitarian 
Assistance Without Host State Consent? Territorial Integrity, Necessity and 
the Determinative Function of the General Assembly’ in: Gill, T.D., Geiß, 
R., Krieger, H., and Mignot-Mahdavi, R. (eds.) Yearbook of International 
Humanitarian Law, Vol. 23. The Hague, T.M.C. Asser Press, pp. 85–121.
Barber, R. (2023) ‘There wasn’t before, and now there even more definitely isn’t, any 
legal barrier to providing cross-border humanitarian assistance in northwest 
Syria’, EJIL:Talk!, 13 February 2023, <https://www.ejiltalk.org/there-wasnt 
-before-and-now-there-even-more-definitely-isnt-any-legal-barrier-to-provi 
ding-cross-border-humanitarian-assistance-in-northwest-syria/> (accessed 20 
April 2023).
BBC (2020) ‘Palantir: The controversial data firm now worth £17bn’, <https://www.bbc.
com/news/business-54348456> (accessed 20 April 2023).
Beduschi, A. (2022) ‘Harnessing the potential of artificial intelligence for humanitari-
an action: Opportunities and risks’ International Review of the Red Cross 104 
(919), pp. 1149–1169.
Buchan, R. (2021) Cyber Espionage and International Law. Hart Publishing.
Businesswire (2022) ‘Palantir Ranked No. 1 in Worldwide Artificial Intelligence Software 
Study in Market Share and Revenue’, <https://www.businesswire.com/news/
home/20220920006178/en/Palantir-Ranked-No.-1-in-Worldwide-Artificial 
-Intelligence-Software-Study-in-Market-Share-and-Revenue> (accessed 20 
April 2023).
Dawidowicz, M. (2017) Third-Party Countermeasures in International Law. Cambridge, 
Cambridge University Press.
250
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review, voL. LXXXiv, 2024
Delerue, F. (2020) Cyber Operations and International Law. Cambridge, Cambridge 
University Press.
European Data Protection Board (2022), Guidelines 05/2022 on the use of facial reco-
gnition technology in the area of law enforcement, <https://edpb.europa.eu/
system/files/2022-05/edpb-guidelines_202205_frtlawenforcement_en_1.
pdf> (accessed 20 April 2023).
FRA (2020) Facial recognition technology: fundamental rights considerations in the 
context of law enforcement, <https://fra.europa.eu/sites/default/files/fra_up 
loads/fra-2019-facial-recognition-technology-focus-paper-1_en.pdf> (acces-
sed 20 April 2023).
Frowein, J. (1987) ‘Collective enforcement of international obligations’ 47 Zeitschrift 
für ausländisches öffentliches Recht und Völkerrecht, pp. 67–79.
Gaja, G. (2011) The Protection of General Interests in the International Community, The 
Hague Academy Collected Courses Online / Recueil des cours de l’Académie 
de La Haye en ligne, Vol. 364, (Martinus Nijhoff Publishers).
Gal, T. (2017) ‘Territorial Control by Armed Groups and the Regulation of Access to 
Humanitarian Assistance’, Israel Law Review 50(1), pp. 25–47.
Gillard, E.-C. (2013) ‘The Law Regulating Cross-Border Relief Operations’, International 
Review of the Red Cross 95 (890), pp. 351–382.
Henckaerts, J.M., (2005) ‘Study on Customary International Humanitarian Law: A 
contribution to the understanding and respect for the rule of law in armed 
conflict’, International Review of the Red Cross, Vol. 87, No. 857, March 2005.
Henckaerts, J.-M., & Doswald-Beck, L. (2005) Customary International Humanitarian 
Law, Volume I, Rules, Cambridge, Cambridge University Press.
ICRC (2014) ICRC Q&A and lexicon on humanitarian access, 96 International Review 
of the Red Cross, No. 893, pp. 359–375.
Jamnejad, M., & Wood, M., ‘The principle of non-intervention’ Leiden Journal of 
International Law 22.02 (2009), pp. 345–381.
Jennings, R., & Watts, A. (eds.) (2008) Oppenheim’s International Law, Ninth edition, 
Oxford, Oxford University Press.
Kop, M. (2021), EU Artificial Intelligence Act: The European Approach to AI, 
Transatlantic Antitrust and IPR Developments.
Kuner, C. (2019) ‘International Organizations and the EU General Data Protection 
Regulation: Exploring the Interaction between EU Law and International Law’, 
International Organizations Law Review, 2019, Vol. 16, No. 1, pp. 158–191.
251
Maruša T. Veber – Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
Kuner, C., & Marelli, M. (2020) Handbook on Data Protection in Humanitarian 
Action, Second edition, ICRC, <https://www.icrc.org/en/data-protection-hu 
manitarian-action-handbook> (accessed 20 April 2023).
Landgren, K., Redhal, G., Romita, P., & Thompson S.K. (2023) ‘The Demise of the 
Syria Cross-Border Aid Mechanism’, Lawfare, 23 August 2023, <https://www.
lawfaremedia.org/article/the-demise-of-the-syria-cross-border-aid-mechani 
sm> (accessed 20 April 2023).
Lindblom, A.-K. (2009) Non-Governmental Organisations in International Law, Cam-
bridge University Press.
Macdonald, A. (2022) ‘African nations must implement safeguards against humanitarian 
digital ID risks: researcher’, Biometric update, <https://www.biometricupdate.
com/202209/african-nations-must-implement-safeguards-against-humanita 
rian-digital-id-risks-researcher> (accessed 20 April 2023).
Martin, A., Sharma G., de Souza S.P., Taylor, L., van Eerd, B., McDonald, S. M., Marelli, 
M., Cheesman, M., Scheel, S., & Dijstelbloem, H. (2023) ‘Digitisation and 
Sovereignty in Humanitarian Space: Technologies, Territories and Tensions’, 
Geopolitics, Vol. 28, No. 3, pp. 1363–1397.
Narbel, V.G., & Sukaitis, J. (2021) ‘Biometrics in humanitarian action: a delicate 
balance’, Humanitarian Law & Policy, <https://blogs.icrc.org/law-and-poli 
cy/2021/09/02/biometrics-humanitarian-delicate-balance/> (accessed 20 
April 2023).
Oxford Guidance on the Law Relating to Humanitarian Relief Operations in Situations 
of Armed Conflict (2016), <https://www.elac.ox.ac.uk/wp-content/uploads/ 
2022/06/oxfordguidancepdfpdf.pdf> (accessed 20 April 2023).
Parker, B. (2019) ‘New UN deal with data mining firm Palantir raises protection con-
cerns’, The New Humanitarian, <https://www.thenewhumanitarian.org/
news/2019/02/05/un-palantir-deal-data-mining-protection-concerns-wfp> 
(accessed 20 April 2023).
Reuters (2019) ‘U.N. food chief warns aid suspension in Yemen likely to start this week’, 
<https://www.reuters.com/article/us-yemen-security-un/u-n-food-chief 
-warns-aid-suspension-in-yemen-likely-to-start-this-week-idUSKCN1TI1 
X7> (accessed 20 April 2023).
Rottensteiner, C. (1999) ‘The denial of humanitarian assistance as a crime under in-
ternational law’, International Review of the Red Cross, Vol. 81, No. 835, pp. 
555–582.
Ryngaert, C., (2013) ‘Humanitarian Assistance and the Conundrum of Consent: A 
Legal Perspective’, Amsterdam Law Forum, 5(2), pp. 5–19.
252
Zbornik znanstvenih razprav – letnik LXXXIV, 2024
LjubLjana Law Review, voL. LXXXiv, 2024
Sancin, V., (ed.) (2010) Lokalni zločinci univerzalni zločini: Odgovornost zaščititi. GV 
Založba, Ljubljana.
Sandoz, Y., Swinarski, C., & Zimmermann, B. (eds.) (1986) Commentary on the 
Additional Protocols of 8 June 1977 to the Geneva Conventions of 12 August 
1949, International Committee of the Red Cross.
Schertel Mendes, L. (2023) The road to regulation of artificial intelligence: the Brazilian 
experience, Internet Policy Review, <https://policyreview.info/articles/news/ro 
ad-regulation-artificial-intelligence-brazilian-experience/1737> (accessed 20 
April 2023).
Schmitt, M. N. (ed.) (2017) Tallin Manual 2.0 on the International Law Applicable to 
Cyber Operations, Prepared by the International Groups of Experts at the 
Invitation of the NATO Cooperative Cyber Defence Centre of Excellence 
(CUP).
Sharpe, M. (2023) Humanitarian access to Gaza, <https://www.ejiltalk.org/humanitari 
an-access-to-gaza/> (accessed 21 February 2024).
Sproson, J., & Olabi, I. (2023) With No Judge or Jury, Who Will Decide the Fate of 4.1 
million Aid-Dependent Syrians? A Comment on the Legality of UN-Coordinated 
Cross-Border Aid Operations in Syria, <https://www.ejiltalk.org/with-no 
-judge-or-jury-who-will-decide-the-fate-of-4-1m-aid-dependent-syrians-a 
-comment-on-the-legality-of-un-coordinated-cross-border-aid-operations-in 
-syria/> (accessed 20 April 2023).
Stoffels, R. A. (2004) ‘Legal regulation of humanitarian assistance in armed conflict: 
Achievements and gaps’, 86 International Review of the Red Cross No. 855, pp. 
515–545.
T. Veber, M. (2022) Sanctions adopted by international organizations in the defence of 
the general interest, PhD Thesis, Faculty of Law, University of Ljubljana.
T. Veber, M. (2023) ‘Z umetno inteligenco podprta humanitarna pomoč in odgovornost 
zaščititi’ Pravna praksa, 29 June 2023, Vol. 42/1597, No. 25, pp. 14–15.
T. Veber, M. (2024) ‘AI-Supported Humanitarian Aid and the Right to Life: Highlighting 
Some of the Legal Challenges Faced by International Humanitarian 
Organizations’ in: Sancin, V. (ed.), Artificial Intelligence and Human Rights: 
From the Right to Life to Myriad of Diverse Human Rights Implications. 
Ljubljana, Litteralis (forthcoming).
T. Veber, M. (2025) ‘International Organizations and AI-Supported Humanitarian 
Aid: Navigating Through the Applicable (Data Protection) Legal Regimes’ 
International and Comparative Law Review (forthcoming).
Tams, C.J. (2005) Enforcing Obligations Erga Omnes in International Law (CUP).
253
Maruša T. Veber – Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
The Guardian (2014) There is no legal barrier to UN cross-border operations in Syria, 
28 April 2014, <https://www.theguardian.com/world/2014/apr/28/no-legal 
-barrier-un-cross-border-syria> (accessed 20 April 2023).
There is Still No Legal Barrier to UN Cross-Border Operations in Syria Without a 
UN Security Council Mandat (2023), <https://www.crossborderislegal.org> 
(accessed 20 April 2023).
UNESCO (2021) Recommendation on the Ethics of Artificial Intelligence, <https://
unesdoc.unesco.org/ark:/48223/pf0000380455> (accessed 1 January 2024).
Welsh, T. (2019) ‘Biometrics disagreement leads to food aid suspension in Yemen’, devex, 
<https://www.devex.com/news/biometrics-disagreement-leads-to-food-aid 
-suspension-in-yemen-95164> (accessed 20 April 2023).
WFP (2016) WFP Introduces Iris Scan Technology To Provide Food Assistance To 
Syrian Refugees, in: Zaatari, <https://www.wfp.org/news/wfp-introduces-in 
novative-iris-scan-technology-provide-food-assistance-syrian-refu> (accessed 
20 April 2023).
Wills, T. (2019) Sweden: Rogue algorithm stops welfare payments for up to 70,000 
unemployed, Algorithm Watch <https://algorithmwatch.org/en/rogue-algori 
thm-in-sweden-stops-welfare-payments/> (accessed 20 April 2023).
Wright, Q. (1962) Espionage and the Doctrine of Non-Intervention in Internal Affairs. 
Essays on Espionage and International Law, pp. 793–798.
353
Zbornik znanstvenih razprav – letnik LXXXIV, 2024 – povzetki
LjubLjana Law Review, voL. LXXXiv, 2024 – synopses
Znanstveni članek
DOI: 10.51940/2024.1.213-249
UDK: 341.33/.34:341.232:004.8
Maruša T. Veber
Umetna inteligenca in humanitarna pomoč: preučitev 
vloge soglasja držav
Avtorica preučuje vlogo in pojem soglasja držav pri zagotavljanju humanitarne pomoči, 
ki jo podpirajo sistemi umetne inteligence, z vidika veljavnih mednarodnih pravil: splo-
šnega pravnega režima, ki ureja humanitarno pomoč, in posebnih pravil, ki izhajajo iz 
mednarodnega prava človekovih pravic ter mednarodnega prava oboroženih spopadov. 
Avtorica ugotavlja, da ima pojem soglasja zadevne države osrednjo vlogo v teh pravilih, 
pri čemer razlikuje med strateškim soglasjem in operativnim soglasjem za zagotovitev 
humanitarne pomoči. Strateško soglasje se nanaša na splošno soglasje države za zago-
tavljanje humanitarne pomoči na njenem ozemlju, operativno soglasje pa se nanaša na 
soglasje, ki se zahteva na operativni oziroma tehnični ravni za zagotavljanje posamezne 
vrste humanitarne pomoči na geografsko opredeljenem območju. Avtorica zatrjuje, da 
je treba utemeljene razloge za zavrnitev operativnega soglasja za humanitarno pomoč, ki 
jo podpira UI in kot izhajajo iz mednarodnega prava oboroženih spopadov, razlikovati 
od samovoljne zavrnitve strateškega soglasja. V prvem primeru je zavrnitev operativnega 
soglasja lahko pravno upravičena, samovoljna zavrnitev strateškega soglasja za dostavo 
humanitarne pomoči pa je prepovedana. Zagotavljanje humanitarne pomoči brez soglas-
ja zadevne države je pravno lahko upravičeno bodisi na podlagi dovoljenja Varnostnega 
sveta Združenih narodov bodisi na podlagi sekundarnih pravil mednarodnega prava, 
zlasti s pravili o protiukrepih.
Ključne besede
umetna inteligenca, humanitarna pomoč, samovoljna zavrnitev soglasja, protiukrepi.
354
Zbornik znanstvenih razprav – letnik LXXXIV, 2024 – povzetki
LjubLjana Law Review, voL. LXXXiv, 2024 – synopses
Scientific Article
DOI: 10.51940/2024.1.213-249
UDC: 341.33/.34:341.232:004.8
Maruša T. Veber
Artificial Intelligence and Humanitarian Assistance: 
Reassessing the Role of State Consent
Abstract
The author analyses the notion of State consent in the delivery of humanitarian assis-
tance supported by artificial intelligence (AI) systems from the perspective of the existing 
applicable international legal regimes, in particular, the general legal regime of human-
itarian assistance and the specific rules deriving from international humanitarian law 
and international human rights law. She argues that the notion of consent lies at the 
heart of these rules with a distinction made between strategic and operational consent to 
humanitarian assistance. The former refers to a State’s general consent to the delivery of 
humanitarian assistance on its territory, while the latter refers to the consent required at 
the operational level for the delivery of a particular type of humanitarian assistance in a 
specific geographically defined area. It is argued that valid reasons for withholding opera-
tional consent to AI-supported humanitarian assistance under international humanitari-
an law must be distinguished from the arbitrary withholding of strategic consent. While 
withholding operational consent may be legally justified, the arbitrary withholding of 
strategic consent to humanitarian assistance is prohibited under the relevant interna-
tional legal regimes when it amounts to a violation of other existing obligations of the 
State concerned (e.g., under international humanitarian law or human rights law). In 
such situations the non-consensual delivery of humanitarian assistance could be legally 
justified either through United Nations Security Council authorisation or by secondary 
rules of international law, in particular countermeasures.
Key words
artificial intelligence, humanitarian assistance, arbitrary withholding of consent, coun-
termeasures.