United Nations Convention on the Prohibition of the Development, Production and Use of Autonomous Weapons Systems

Introduction

At the beginning of the 21st century, humanity faced qualitatively new challenges to global security due to the rapid development of artificial intelligence and robotics technologies in the military sphere. As Kochetkov (2010) rightly points out, changes in approaches to international security at the present stage require a rethinking of traditional legal and ethical paradigms. In this regard, the issue of autonomous weapons systems is of particular interest, since, according to many researchers, they may radically transform the nature of armed conflicts and the very nature of war (Baklanov, 2013; Frolov, 2023). Moreover, as noted in the scientific literature, the integration of artificial intelligence into military technologies poses unprecedented risks to strategic stability and the international law system (Sharikov, 2018; Karasev, 2020).

First of all, it is noteworthy that autonomous weapons systems represent a qualitatively new class of military technologies capable of independently selecting and hitting targets without direct human involvement in deciding whether to use force (Scharre, 2022). Researchers believe that this raises fundamental questions about the compatibility of such systems with the basic principles of international humanitarian law, including the principles of distinction, proportionality and military necessity (Winter, 2022; Ahmed et al., 2026). As Gunawan (2022) showed, the use of autonomous weapons poses serious problems for the implementation of the concept of command responsibility under international criminal law, as the traditional chain of responsibility between the operator, commander and system developer is blurring. One has to emphasize that the issue of legal responsibility for the actions of autonomous systems remains one of the most complex and unresolved in modern legal doctrine (Passinhas, 2018; Baranov, 2020; Miller, 2025).

Analysis of the materials indicates that the development and implementation of combat robots of various classes, from unmanned aerial vehicles to ground-based combat platforms, are becoming global (Yamauchi, 2004; Bhargavi & Manjunath, 2011). The most noticeable progress is in the field of swarm intelligence used in military drones, as was demonstrated in military conflicts (King, 2024). At the same time, humanoid combat robots and autonomous platforms for urban military operations are being actively developed (Uthai et al., 2026). It is very significant that modern military robotics systems are integrated with the Internet of Things, creating comprehensive surveillance and automated response networks (Gopal & Lokesh, 2025; Dhayal et al., 2026). Such systems are used not only in traditional military operations, but also in countering terrorism, as evidenced by Europol materials (Liu et al., 2025). Even in the educational field, there is a trend to introduce combat robots into military training programs for young people, which reflects the systemic militarization of robotics (Zhao et al., 2023).

At the same time, the international legal regulation of autonomous weapons systems remains fragmented and insufficient. As noted in the works by Arbatov (2019) and Mizin (2023), the traditional arms control system faces a deep crisis and needs to adapt to new technological realities. It is characteristic that states demonstrate different approaches to the need to ban or restrict autonomous weapons (Qerimi, 2023). According to the typology of state positions presented by Qerimi (2023), they range from a complete ban to the absence of any restrictions on the autonomous weapons development. As Agarwal and Singh (2020) show, India takes an intermediate position in this discussion, recognizing the need for certain legal mechanisms while maintaining technological development. It should be noted that the use of autonomous systems in the Ukrainian conflict has raised issues of compliance of such weapons with the norms of international humanitarian law (Mewoh & Rahmadan, 2025).

The conducted studies have shown the existence of significant technical risks associated with the operation of lethal autonomous weapons systems (Podar & Colijn, 2025). According to researchers, these risks include the unpredictable behavior of artificial intelligence systems in unusual situations, vulnerability to cyber attacks, and the possibility of uncontrolled escalation of conflicts. In this regard, characteristic is the statement by Golan (2023) about the need to move from the cybersecurity concept to that of cyber-protection in the context of artificial intelligence systems. It is significant that the integration of artificial intelligence into the military sphere creates new vectors of threats to strategic stability, since autonomous systems can be used for both conventional and nuclear weapons (Sokova, 2020; Stefanovich, 2025). As Sharikov (2018) rightly points out, the combination of artificial intelligence, cyberattacks, and nuclear weapons is dangerous and may destabilize the global security system.

In addition, the militarization of artificial intelligence systems generates a set of ethical problems that go beyond the exclusively legal regulation. It is noteworthy that entitling machines to make decisions about human life and death contradicts fundamental humanistic values and the principle of human dignity (Reichberg & Syse, 2021; Gnatik, 2024). Although some authors, such as Müller (2016), speculate about the potentially positive aspects of autonomous weapons systems, the vast majority of researchers emphasize their threats to humanity. The following points should be highlighted: autonomous weapons systems can transform war into a technical process devoid of moral responsibility; they create the risk of lowering the threshold for entering into armed conflict; they increase the asymmetry between technologically advanced and developing countries (Ruschi, 2020; Rosert & Sauer, 2021).

Along with military-political and ethical-legal aspects, we must verify the fact that the development of military robotics and autonomous systems has significant socio-economic consequences. Research in the field of automation economics indicates the need for government regulation of robotic systems, including discussions on their taxation (Zhang, 2019; Thuemmel, 2023; Carbonara et al., 2024). In relation to the military sphere, this issue is becoming particularly acute, since the weapons automation affects the labor market in the defense industry and changes the employment structure in the armed forces (Prettner & Strulik, 2020; Costinot & Werning, 2023). It should also be noted that issues of technological confrontation and ensuring military security under the developing convergent technologies require an integrated approach on the part of states (Pankova and Gusarova, 2019; Strelnikov, 2022; Stefanovich, 2025).

As Istyagin (2017) quite rightly notes, a global solution to the problems associated with new weapons technologies is an imperative condition for maintaining international security. In the context of autonomous weapons systems, this means the need to form a universal international legal regime based on a multilateral treaty. As is well known, the mechanism for developing such regimes within the United Nations demonstrated its effectiveness in relation to other categories of weapons that had aroused particular concern of the international community (Annan, 2005). Now let us study the campaigns for the prohibition of certain types of weapons. A comparative analysis of the movements for nuclear disarmament and the prohibition of autonomous weapons conducted by Rosert and Sauer (2021) shows both general patterns and specific features of these processes. At the same time, one should admit that the current arms control crisis poses serious obstacles to achieving international consensus (Brooks et al., 2022; Bogdanov & Yevtodyeva, 2022).

Nevertheless, the analysis of existing legal problems, technological risks and ethical challenges convincingly proves the need to adopt a special international treaty. It follows from the above that such a treaty should contain clear definitions of autonomous weapons systems, criteria for an acceptable level of autonomy, requirements for maintaining meaningful human control over the use of force, and verification and international monitoring mechanisms. Summing up the scientific literature analysis, one can state that the development and implementation of the United Nations Convention on the Prohibition of the Development, Production and Use of Autonomous Weapons Systems is an urgent necessity at the current stage of the international law and global security system development.

1. Explanatory notes to United Nations Convention on the Prohibition of the Development, Production and Use of Autonomous Weapons Systems

The United Nations Convention on the Prohibition of the Development, Production and Use of Autonomous Weapons Systems, the first in the history of international law, was presented to the attention of the diplomatic corps, representatives of international organizations, government agencies, the scientific community, non-governmental organizations and all interested parties. It is a fundamental international legal document designed to prevent potentially irreversible transformations in the nature of warfare.

The transformational processes in the field of military technology, characterized by the rapid development of artificial intelligence and robotics systems, have created an unprecedented challenge to the international legal order, requiring a consolidated response from the world community. As is known, the convergence of advanced technologies and the military-industrial complex provide the technical possibility of developing autonomous weapons systems operating without meaningful human control. It is interesting in this regard that this circumstance opens up the prospect of a fundamental transformation of the nature of warfare with unpredictable consequences for international security and humanitarian law.

First of all, it is notable that the Convention forms the conceptual and terminological basis for the international legal regulation of autonomous weapons systems. In particular, it defines the parameters of meaningful human control as a fundamental element of the legitimate use of force and establishes absolute prohibitions on the development and production of fully autonomous weapons systems. In addition, it introduces the obligations of states to destroy existing systems, creates an institutional implementation mechanism in the form of an Organization for the prohibition of autonomous weapons systems, and provides effective procedures for verifying compliance with the Convention provisions.

One should mind that the developed Convention is based on a systematic and integrated approach to the preventive regulation of military technologies. It is aimed at forming an international legal regime that ensures the preservation of human control over the use of force as a fundamental principle of international humanitarian law. In other words, the Convention must prevent a qualitative change in the nature of armed conflicts, when decisions about life and death are delegated to autonomous technical systems.

The proposed Convention is an international legally binding instrument aimed at establishing a comprehensive preventive ban on autonomous weapons systems. It should be noted that the document is organically integrated into the existing architecture of international humanitarian law, as well as the fundamental principles of international law relating to human rights and international security.

Moreover, the Convention complies with the United Nations Charter¹, the Geneva Convention and its relevant annexed Protocols², the Convention on Prohibitions or Restrictions on the Use of Certain Conventional Weapons Which May Be Deemed to Be Excessively Injurious or to Have Indiscriminate Effects³, as well as the fundamental principles of international law, human rights, and international security. It is essential that the document creates the necessary legal conditions to prevent a potentially destabilizing arms race and minimize the risks of uncontrolled escalation of conflicts, which is confirmed by the historical experience of preventive regulation of other types of mass destruction weapons.

It is characteristic that the theoretical base of the Convention relies on the conceptual provision that it is inadmissible to delegate decisions on the lethal force use to technical systems operating outside the limits of meaningful human control. In our opinion, such delegation would inevitably lead to an unacceptable gap in the chain of legal responsibility and would contradict the fundamental principles of distinction, proportionality and precautions, which form the normative basis of international humanitarian law.

We take the liberty to assert that delegating decisions on the use of lethal force to autonomous systems is not only a legal, but also a deep ethical problem affecting the very nature of human responsibility and moral subjectivity in the context of armed conflicts. That is why the Convention proceeds from the imperative requirement of maintaining meaningful human control at all stages of using force.

The Convention structure provides for a clear articulation of the object and subject composition of regulation. The central element is the legal definition of “autonomous weapons systems” as systems capable of independently identifying, selecting and hitting targets without meaningful human control after their activation. At the same time, the Convention establishes a criterion framework for differentiating prohibited autonomous systems and permissible automated systems through the concept of “meaningful human control” characterized by information sufficiency, possible effective intervention, and established responsibility.

The following points should be highlighted: the concept of meaningful human control presupposes not just the formal presence of a person in the control loop, but the actual ability of the operator to understand the context of the system application, make an informed decision and effectively intervene in the process of selecting and defeating targets. Thus, the Convention establishes a functional rather than a technical criterion for distinguishing between permissible and prohibited systems.

The most important thing is that the regulatory system of obligations of the participating states presupposes an absolute ban on the development, production, acquisition, transfer and use of autonomous weapons systems operating without meaningful human control. These obligations are complemented by the imperative requirement to destroy existing systems and implement effective national measures to ensure compliance with the Convention.

In this regard, we must add that the Convention provides for a phased mechanism for fulfilling obligations, including the declaration of existing development programs, the suspension of all work on the creation of prohibited systems, and their subsequent destruction under international control. At the same time, the participating states undertake not to assist, encourage or force other persons to carry out any activity prohibited by the Convention.

One should also note that the institutional mechanism for the Convention implementation is the ad hoc Organization for the Prohibition of Autonomous Weapons Systems, whose competence covers compliance monitoring, technical expertise, and the promotion of international cooperation. At the same time, the Convention verification regime is based on a multi-component system that includes regular declarations, inspections, and special investigations of alleged violations.

At the same time, the Organization functions include coordinating international cooperation in the development of artificial intelligence and robotics technologies for peaceful purposes, facilitating the exchange of scientific and technical information and providing technical assistance to the participating states in the implementation of the Convention. It is essential that the institutional architecture of the Organization ensures a balance between the effective verification and respect for the legitimate interests of states in the field of national security.

It is extremely important that the adoption of the Convention will create the necessary legal conditions to prevent a potentially destabilizing arms race in the field of autonomous systems and minimize the risks of uncontrolled escalation of armed conflicts. As the history of regulating chemical, biological and nuclear weapons shows, preventive regulation helps to avoid the much more complex and resource-intensive task of controlling systems already widespread.

In this regard, it is typical to say that preventing the proliferation of dangerous weapons at the early stages of their technological development is a more effective and less costly approach than attempts to limit systems already deployed. This leads to the conclusion that it was strategically important to adopt the Convention timely, before the autonomous weapons systems reach the stage of mass production and deployment.

It is fair to note that the Convention does not limit the development of artificial intelligence and robotics technologies for peaceful and defense purposes, provided that meaningful human control over the use of force is maintained. This ensures a balance between national security imperatives and fundamental humanitarian values, allowing states to continue technological development in the field of defense without creating unacceptable risks to international security and compliance with humanitarian law.

In other words, the Convention makes a clear distinction between prohibited fully autonomous weapons systems and permissible automated systems operating under meaningful human control. This approach allows preserving technological progress in the defense sector, while simultaneously eliminating the most dangerous and morally unacceptable applications of artificial intelligence for military purposes.

This paper is intended for diplomats, international lawyers, representatives of the legislative and defense sectors, researchers, professors and students, as well as a wide range of readers interested in international security, humanitarian law, and ethical aspects of military technology. It can serve as a basis for diplomatic negotiations, national lawmaking, scientific research and educational programs in the field of international law and technological ethics.

Summing up, one can state that the proposed Convention is the most important tool of preventive international legal regulation aimed at ensuring humane and sustainable technological development in the interests of maintaining international peace and security of the humanity. Thus, its adoption is not only a legal necessity to fill the existing gap in international law, but also a moral imperative to preserve human dignity and fundamental humanitarian values in the era of rapid technological advancement.

All this convincingly proves the relevance and timeliness of the development of this international legal instrument, which is designed to ensure that technological development in the military sphere remains under human control and serves the purpose of protecting, rather than threatening, fundamental human rights and freedoms.

2. Structure of the United Nations Convention on the Prohibition of the Development, Production and Use of Autonomous Weapons Systems

Preamble

Section I. General provisions

Article 1. General obligations

Article 2. Definitions and criteria

Article 3. Exceptions

Section II. Implementation and verification

Article 4. National implementation policies

Article 5. Declaration and destruction

Article 6. Organization for the Prohibition of Autonomous Weapons Systems

Article 7. Control and verification of compliance

Section III. Cooperation and assistance

Article 8. Consultations, cooperation and fact-finding

Article 9. International cooperation and technical assistance

Article 10. Promoting the development of control technologies

Section IV. Other provisions

Article 11. Liability for violations

Article 12. Relation to other international treaties

Article 13. Measures to eliminate violations and ensure compliance

Article 14. Terms of duration and withdrawal from the Convention

Article 15. Status of Annexes

Article 16. Amendments

Article 17. Registration

Article 18. Authentic texts

Annex I. Technical criteria for defining autonomous weapon systems

Annex II. Inspection and verification procedures

Annex III. Confidentiality of information

3. United Nations Convention on the Prohibition of the Development, Production and Use of Autonomous Weapons Systems

Preamble

The States Parties to this Convention,

Recognizing the devastating consequences of armed conflicts for humanity and the need to prevent such suffering in the future,

Concerned that the rapid development of artificial intelligence, robotics and autonomous systems technologies poses unprecedented threats to international peace, security and stability,

Taking into account the principles and norms of international humanitarian law enshrined in the 1949 Geneva Convention and its annexed Protocols, the 1907 Hague Convention, and other relevant international instruments,

Reaffirming the martens clause, according to which, in cases not provided for by existing agreements, the inhabitants and the belligerents remain under the protection and the rule of the principles of the law of nations, as they result from the usages established among civilized peoples, from the laws of humanity, and the dictates of the public conscience,

Convinced that the conscience of humanity would be deeply shocked by the idea that decisions about life and death can be made by machines without meaningful human control and responsibility,

Recognizing that autonomous weapons systems pose serious ethical, legal, military and diplomatic challenges, including the risks of unpredictable behavior, escalation of conflicts, an arms race and lowering the threshold for the use of force,

Emphasizing that the rules and principles of international humanitarian law, including the principles of distinction, proportionality and precaution, require a contextual understanding and legal assessment that cannot be fully delegated to technical systems,

Recognizing that international law establishes personal responsibility for serious violations of international humanitarian law and that such responsibility cannot be transferred to autonomous systems,

Taking into account the relevant decisions and recommendations of the united nations general assembly, the united nations security council, the conference on disarmament and other international bodies concerning new technologies in the field of armaments,

Welcoming the efforts made by States, international organizations, the International Committee of the Red Cross, non-governmental organizations and civil society to address humanitarian and international security issues that may be caused by autonomous weapons systems,

Determined to act in order to achieve effective progress towards general and complete disarmament under strict and effective international control,

Desiring to promote the implementation of the purposes and principles of the Charter of the United Nations,

Reaffirming the right of each State to the peaceful use of advanced technologies and the obligation of all States to cooperate in establishing an effective system of international control guaranteeing the peaceful use of such technologies,

Have agreed as follows:

Section I

General provisions

Article 1

General obligations

1. Each State Party to this Convention undertakes under any circumstances:

a) not to develop, manufacture, otherwise acquire, accumulate, preserve or transfer autonomous weapons systems to anyone, directly or indirectly;

b) not to use autonomous weapons systems under any circumstances;

c) not to assist, encourage or in any way facilitate anyone to engage in activities prohibited by the States Parties to this Convention;

d) not to use advances in digital technology and artificial intelligence to create systems functionally equivalent to autonomous weapons systems.

2. Each State Party undertakes to destroy or modify all autonomous weapons systems that it possesses or that are under its jurisdiction or control, in accordance with the provisions of this Convention.

3. Each State Party undertakes to ensure that any weapons system under its jurisdiction or control operates under meaningful human control.

Article 2

Definitions and criteria

For the purposes of this Convention:

1. “Autonomous weapon system” means any weapon system that is capable of selecting (i.e., searching, identifying, tracking, selecting) and hitting (i.e., using force, neutralizing, damaging, or destroying) targets without meaningful human control and regardless of the degree of technical complexity.

2. “Meaningful human control” means the level of human intervention at which people:

a) possess sufficient information and understanding of the functioning of the system and the operational environment;

b) maintain a consistent ability to make informed decisions about the use of force, including timely intervention and deactivation;

c) are responsible for making decisions regarding the use of lethal force;

d) can actively monitor the functioning of the system throughout its application.

3. “Development” means all stages prior to industrial production, such as design, research, testing, prototyping, modification of existing systems, and other preparatory activities.

4. “Manufacturing” means all stages of manufacture, including design, manufacture, integration, assembly, testing, storage and related activities.

5. “Acquisition” means obtaining by any means, including purchase, leasing, loan, technology transfer, or any other form of access to systems or their components.

6. “Weapon system” means a warfare means capable of causing damage, harm or destruction, which may include a platform, target detection equipment, control systems, and damaging elements per se.

7. “Component of an autonomous weapon system” means any part, element, or subsystem specifically designed to function as an integral part of an autonomous weapon system.

Article 3

Exceptions

1. The following systems are not autonomous weapons systems for the purposes of this Convention:

a) anti-missile, anti-aircraft and anti-artillery defense systems designed solely to protect against inanimate objects (missiles, projectiles, air targets), provided they are under meaningful human control;

b) systems that operate for a limited time and in a limited geographical area under the continuous supervision and control of an operator capable of terminating their operation;

c) mine clearance, search and disposal systems for explosive objects used exclusively for protective and humanitarian purposes;

d) systems that use automation only for moving, navigating, or self-locating functions, but not for selecting and defeating targets;

e) training systems without combat capabilities;

f) non-military systems used for law enforcement, civil, humanitarian, scientific and educational purposes.

2. The Participating States undertake to ensure that systems subject to exceptions cannot be modified to function as autonomous weapons systems.

Section II

Implementation and verification

Article 4

National implementation policies

1. Each State Party shall, in accordance with its constitutional procedures, take the necessary measures to prohibit and prevent any activity prohibited to a State Party to this Convention, carried out anywhere under its jurisdiction or control.

2. Each State Party shall adopt legislative, administrative, judicial and other measures, including the establishment of criminal liability, in order to implement this Convention and comply with its provisions.

3. Each State Party shall prohibit natural and legal persons located on its territory or under its jurisdiction from carrying out any activity prohibited for the States Parties to this Convention.

4. Each State Party shall establish or designate a national authority that shall act as a national focal point for effective communication with other Participating States and the Organization for the Prohibition of Autonomous Weapons Systems. The State Party shall notify the Organization of its national authority upon accession to this Convention.

5. Each State Party shall inform the Organization of the legislative and administrative measures taken to implement this Convention.

Article 5

Declaration and destruction

1. Each State Party, no later than 30 days after the entry of this Convention into force for it, shall declare to the Organization:

a) whether it possesses or owns or has autonomous weapons systems in any territory under its jurisdiction or control;

b) whether it has transferred or received, directly or indirectly, autonomous weapons systems, as well as specifications on any program for the development or production of such systems that it had in place at the time of entry of this Convention into force for it.

2. Each State Party shall, no later than 120 days after the entry of this Convention into force for it, provide the Organization with detailed plans for the destruction or modification of autonomous weapons systems owned or possessed by it, or available in any territory under its jurisdiction or control.

3. All autonomous weapon systems referred to in Paragraph 1 of this Article shall be destroyed or modified no later than 8 months after the entry of this Convention into force, in accordance with the announced destruction or modification plan.

4. Each State Party, during the destruction of autonomous weapons systems, shall ensure the safety of people and the protection of the environment.

5. Each State Party shall cooperate with other Participating States in the exchange of equipment, materials, scientific and technological information on the destruction of autonomous weapons systems.

Article 6

Organization for the Prohibition of Autonomous Weapons Systems

1. The States Parties to this Convention shall establish the Organization for the Prohibition of Autonomous Weapons Systems (hereinafter referred to as the “Organization”) to achieve the objective and goal of this Convention, to ensure the implementation of its provisions, including those relating to international verification of compliance with the Convention, and to provide a forum for consultations and cooperation between the Participating States.

2. The organs of the Organization are:

a) The Conference of the Participating States;

b) The Executive Board;

c) The Technical Secretariat.

3. The Conference of the Participating States (hereinafter referred to as “the Conference”) shall consist of all Participating States to the Convention. Each State Party shall have one representative at the Conference, who may be accompanied by alternates and advisers.

4. The first session of the Conference shall be convened by the depositary not later than 30 days after the entry into force of this Convention.

5. The Conference shall meet in regular sessions, which shall be held annually, unless it decides otherwise.

6. The Executive Board (hereinafter referred to as “the Board”) shall consist of 41 members. Each Participating State has the right, in accordance with the principle of rotation, to be a Board member. The Board members are elected by the Conference for a two-year term.

7. The Board is the executive body of the Organization. It shall be accountable to the Conference and perform the powers and functions assigned to it by the present Convention.

8. The Technical Secretariat (hereinafter referred to as “the Secretariat”) shall assist the Conference and the Board in the performance of their functions. The Secretariat shall carry out the verification measures provided for in the present Convention.

Article 7

Control and verification of compliance

1. In order to verify compliance with the provisions of this Convention, each State Party agrees to monitoring, including:

a) national declarations and reporting;

b) scheduled inspections of facilities;

c) on-demand inspections in case of suspected non-compliance with this Convention;

d) investigation of the alleged use of autonomous weapons systems.

2. Each State Party shall submit to the Organization annual announcements regarding its activities relevant to the objectives of this Convention, including:

a) up-to-date information on its national legislation and administrative measures taken to implement this Convention;

b) a report on developments in artificial intelligence technologies, robotics and autonomous systems related to the military field;

c) information on any military systems being developed or manufactured that have automation elements but are not autonomous weapons systems as defined in this Convention.

3. The Technical Secretariat shall be authorized to carry out inspections in the territory of the Participating States in order to verify compliance with the provisions of this Convention.

4. Each State Party has the right to request an inspection on request in the territory of any other State Party in order to clarify and resolve any issues regarding possible non-compliance with the provisions of this Convention.

5. Inspection teams shall consist of inspectors and assistant inspectors appointed by the Director General from a list of inspectors and assistant inspectors provided by the Participating States.

6. The inspected State Party is obliged to provide access within the inspected territory for the sole purpose of establishing facts relevant to possible non-compliance with the provisions of this Convention.

Section III

Cooperation and assistance

Article 8

Consultations, cooperation and fact-finding

1. The Participating States agree to consult and cooperate with each other in resolving any issues that may arise with respect to the subject matter of this Convention or the implementation of its provisions.

2. Without prejudice to the right of any State Party to request an inspection upon request, the Participating States should, first of all, make every effort to clarify and resolve, through the exchange of information and consultations, any issue that may raise doubts about compliance with this Convention.

3. A State Party that has received a request from another State Party for clarification of any matter deemed to be in doubt or concern shall provide the requesting State Party with sufficient information to clarify the matter no later than 15 days after receiving the request.

4. If the issue is not resolved, the requesting State Party may request an extraordinary session of the Executive Board.

Article 9

International cooperation and technical assistance

1. The Participating States undertake to facilitate the fullest possible exchange of equipment, materials, scientific and technical information for the effective implementation of this Convention.

2. The Organization shall develop and implement programs of cooperation and technical assistance to the Participating States in order to effectively implement the provisions of the Convention.

3. Nothing in this Convention shall be interpreted as limiting or impeding the right of the Participating States to research, develop, produce, acquire, transfer or use scientific knowledge and technology for peaceful purposes.

4. The Participating States undertake to promote international cooperation in the peaceful use of artificial intelligence and robotics technologies, including scientific and technical exchange.

5. The Technical Secretariat shall establish and maintain a database containing information related to the various means of implementing this Convention, as well as information provided by the Participating States.

Article 10

Promoting the development of control technologies

1. The Participating States undertake to develop and improve verification methods, including the creation of standards and criteria for weapons systems with automation elements, in order to ensure a clear distinction between permissible automation and prohibited autonomous systems.

2. The Participating States undertake to cooperate with the Organization in developing guidelines and protocols for inspections and verification.

3. The Organization shall promote the development of technologies and techniques for effective monitoring and verification of compliance with the provisions of this Convention.

Section IV

Other provisions

Article 11

Liability for violations

1. Each State Party shall, in accordance with its constitutional procedures, take the necessary measures to establish criminal liability of natural and legal persons for violations of the prohibitions established by this Convention.

2. With respect to activities prohibited for a State Party to this Convention, each State Party shall, in accordance with international law, establish its jurisdiction when such activities are carried out:

a) in the territory of that State or in any other place under its jurisdiction or control;

b) by any of its citizens;

c) by natural or legal persons directly or indirectly under its control.

3. Persons responsible for serious violations of this Convention shall be prosecuted, regardless of their nationality and place of the violation commission.

Article 12

Relation to other international treaties

1. Nothing in this Convention shall be interpreted as limiting or detracting from the obligations assumed by any State in accordance with the Charter of the United Nations or existing international treaties in the field of disarmament and arms control.

2. This Convention supplements the existing norms of international humanitarian law and international human rights law.

Article 13

Measures to eliminate violations and ensure compliance

1. The Conference, on the recommendation of the Executive Board, shall take the necessary measures to ensure compliance with this Convention and to eliminate and remedy any situation that is contrary to the provisions of the Convention.

2. In cases of particular seriousness, the Conference shall bring the issue, including relevant information and conclusions, to the attention of the United Nations General Assembly and the United Nations Security Council.

Article 14

Terms of duration and withdrawal from the Convention

1. This Convention is open-ended.

2. Each State Party, in the exercise of its national sovereignty, has the right to withdraw from this Convention if it decides that exceptional circumstances relating to the subject matter of this Convention have jeopardized its supreme interests.

3. Withdrawal shall be effected by giving 90 days notice of withdrawal to all other Participating States, the Executive Board, the depositary, and the United Nations Security Council. Such notification shall indicate the exceptional circumstances that the State Party considers to have jeopardized its supreme interests.

Article 15

Status of Annexes

1. The Annexes are an integral part of this Convention. Any reference to this Convention shall include its Annexes.

Article 16

Amendments

1. Any State Party may propose amendments to this Convention. Any State Party may also propose amendments to the Annexes to this Convention.

2. The text of the proposed amendment shall be submitted to the Director General, who shall transmit it to all Participating States and to the depositary. The proposed amendment is considered only by the Amendment Review Conference. Such an Amendment Review Conference shall be convened if one third or more of the Participating States notify the Director General, no later than 30 days after the circulation of the proposal, of their support for further consideration of the proposal.

3. Amendments shall enter into force for all Participating States 30 days after depositing of instruments of ratification or acceptance by all Participating States referred to in Paragraph 1.

Article 17

Registration

1. This Convention shall be registered by the depositary in accordance with Article 102 of the Charter of the United Nations.

Article 18

Authentic texts

1. This Convention, of which the Arabic, Chinese, English, French, Portuguese, Russian and Spanish texts are equally authentic, shall be deposited with the Secretary-General of the United Nations.

Annex I

Technical criteria for defining autonomous weapon systems

1. Technical characteristics of the system indicating its autonomy:

a) the ability to independently determine the parameters of the use of force (time, place, intensity, duration);

b) the ability to independently identify, classify and select targets without direct guidance from a human operator;

c) the ability to adapt one’s behavior and make decisions based on changes in the operational environment without explicit human reprogramming;

d) the limited ability of a human to understand the logic of real-time decision-making by the system;

e) the inability of the operator to timely intervene in the decision-making process on the use of force.

2. Technical characteristics that ensure meaningful human control:

a) reliable and secure communication channels between the operator and the system;

b) human-understandable interfaces for managing and displaying information about the system functioning;

c) technical means of emergency shutdown and deactivation available to the operator;

d) software and hardware restrictions that prevent the use of force without the authorization of the operator;

e) means of recording and storing data on all system actions for subsequent analysis.

Annex II

Inspection and verification procedures

1. The procedure for conducting inspections of facilities:

a) notification of the inspected State Party;

b) decision on the inspection team;

c) entry into the territory of the inspected State Party;

d) access to inspected facilities;

e) conducting inspection activities;

f) preparation and submission of an inspection report.

2. Rights and obligations of the inspection team and the inspected State Party:

a) the right of the inspection team to access relevant facilities and information;

b) the right of the inspected State Party to protect confidential information unrelated to the goals of the inspection;

c) the obligation of inspectors to carry out activities with minimal interference to the normal functioning of facilities;

d) the obligation of the inspected State Party to provide assistance to the inspection team.

Annex III

Confidentiality of information

1. Principles for the treatment of confidential information:

a) protection of confidential information obtained by the Organization in the course of the implementation of this Convention;

b) ensuring that confidentiality requirements do not impede effective verification of compliance with the Convention;

c) establishing levels of confidentiality of information and appropriate access procedures;

d) liability for breach of confidentiality.

2. Procedure for handling confidential information:

a) collection, processing, storage and transfer of confidential information;

b) access of the Organization’s personnel to confidential information;

c) protection of facilities and confidential data;

d) procedures in case of violation or alleged violation of confidentiality.

In witness whereof, the undersigned, being duly authorized thereto, have signed this Convention.

Committed on “___” ___________ in a single copy in the Arabic, Chinese, English, French, Russian and Spanish languages, all texts being equally authentic.

Conclusions

The presented United Nations Convention on the Prohibition of the Development, Production and Use of Autonomous Weapons Systems is an innovative preventive international legal instrument, designed to prevent a qualitative transformation of the nature of armed conflicts in the era of technological convergence of artificial intelligence and the military sphere.

The conceptual significance of the proposed document lies in the formation of a normative paradigm based on the imperative of maintaining human control over decisions on the use of deadly force. The introduction of the “meaningful human control” category into international legal circulation with clearly defined criteria of information sufficiency, as well as the possibility of effective intervention and establishing responsibility, creates a fundamental basis for differentiating legitimate automated systems and prohibited autonomous weapons systems.

The structural architecture of the Convention provides comprehensive regulation of the entire life cycle of autonomous weapons systems, from a ban on development to an imperative requirement to destroy existing systems within an eight-month period. The creation of a specialized institutional mechanism in the form of an Organization for the Prohibition of Autonomous Weapons Systems and the establishment of a multi-component verification regime guarantee effective implementation and monitoring of compliance with the Convention obligations.

The preventive nature of legal regulation reflects a qualitatively new approach in international disarmament law, which makes it possible to preempt a potentially destabilizing arms race before the mass deployment of appropriate technologies. The historical experience of regulating chemical, biological and nuclear weapons strongly suggests that the establishment of prohibitions at the initial stages of technological development is much more effective than attempts to control already widespread systems.

The proposed document organically integrates into the existing system of international humanitarian law, supplementing the norms of the 1949 Geneva Convention and its annexed Protocols, as well as the 1981 Convention on Prohibitions or Restrictions on the Use of Certain Conventional Weapons. The Convention extends the Martens Clause onto the latest military technologies, confirming that in cases not covered by existing agreements, combatants and civilians remain protected by the principles of international law and the requirements of humanity.

The practical implementation of the Convention provisions will help minimize the risks of uncontrolled escalation of armed conflicts, eliminate gaps in the chain of legal responsibility when using deadly force, and ensure compliance with the fundamental principles of distinction, proportionality, and precaution under the technological transformation of the military sphere.

At the same time, one should note that the effectiveness of the proposed international legal regime will crucially depend on the political will of States to universalize the Convention; on reaching consensus among the leading military powers on the parameters of prohibitions and exceptions; and on the ability of the verification mechanism to adapt to the rapid development of relevant technologies.

The Convention does not limit the development of artificial intelligence and robotics technologies for peaceful and defense purposes, provided that meaningful human control is maintained. This ensures the necessary balance between national security imperatives and fundamental humanitarian values. The document creates the legal basis for responsible technological development, putting a human and their dignity at the center of the military technology evolution.

The adoption of this Convention by the international community would be a historic achievement of preventive diplomacy in the field of disarmament and an essential contribution to ensuring sustainable and humane technological development in the interests of maintaining international peace and human security.