Council of Europe Framework Convention on Artificial Intelligence in the Context of Human Rights Protection: Challenges and Significance for Ukraine

Abstract

The article provides a comprehensive scientific and legal analysis of the Council of Europe Framework Convention on Artificial Intelligence through the lens of the protection of human rights, democratic values, and the principle of the rule of law, as well as to the determination of its significance and potential challenges for the legal system of Ukraine.

The author substantiates the transition from ethical and soft-law approaches to a risk- based model of legally binding regulation of artificial intelligence, which provides for the accountability of algorithmic systems and increased judicial review over automated decisions of public authorities (administrative bodies); the place of the Convention in the system of international and European law, its relationship with the Convention for the Protection of Human Rights and Fundamental Freedoms, the practice of the European Court of Human Rights and modern European standards for the regulation of digital technologies in public administration and in other areas are investigated .

The article pays special attention to the analysis of the fundamental principles of the Council of Europe Framework Convention on Artificial Intelligence, in particular the orientation towards a human-centric approach, ensuring respect for human dignity, non-discrimination, transparency, accountability and prevention of abuse in the process of development, implementation and use of artificial intelligence. The author reveals the content of the legal mechanism of the Convention aimed at minimizing the risks of violation of human rights, in particular the right to privacy (in the broad sense of the right to private life), protection of personal data, freedom of expression, the right to a fair trial and the prohibition of discrimination.

The article argues that the rapid development of artificial intelligence technologies poses new challenges for traditional legal institutions that require proper legal regulation. It is concluded that the Council of Europe Framework Convention on Artificial Intelligence is an attempt to form globally applicable legal framework for the legal regulation of artificial intelligence in order to ensure a balance between digital progress and the effective protection of human rights.

Based on the conducted scientific analysis, the article reveals the significance of the Convention for Ukraine in view of the European integration processes, obligations as a member of the Council of Europe and the prospects for implementing its provisions into national legislation. Legal, institutional and practical challenges and problems associated with the implementation of new standards for regulating artificial intelligence into the Ukrainian legal system are identified. Based on the results of the study, proposals have been developed for further harmonization of national legislation with the provisions of the Framework Convention in order to strengthen human rights guarantees in the digital age.

Keywords: Framework Convention on Artificial Intelligence, Council of Europe, artificial intelligence, rule of law, human rights, European standards, democratic values, digital technologies.

Relevance of the research topic. The rapid development and implementation of artificial intelligence (AI) technologies in all areas of legal relations have led to the emergence of new challenges for traditional legal institutions, in particular in the field of ensuring and protecting human rights, the development of democratic processes and compliance with the rule of law. The use of algorithmic systems in public administration, justice, security, personal data processing and other socially significant areas has led to the need to define clear international legal standards capable of ensuring the proper balance between digital development and legal guarantees for the protection of individual rights. In modern conditions, the problem of algorithmic management in the public sector is particularly relevant, when automated systems are used to make decisions with legally significant consequences that directly affect the scope of the rights and obligations of individuals.

In order to ensure the protection of human rights in the field of AI, on May 17, 2024, at the 133rd meeting of the Committee of Ministers of the Council of Europe, the Framework Convention of the Council of Europe on Artificial Intelligence, Human Rights, Democracy and the Rule of Law (hereinafter referred to as the AI Convention) [1] was adopted, which acquired the status of the first legally binding international document aimed at comprehensive legal regulation of the use of AI. The adoption of this document led to the transition from recommendatory and ethical approaches ("soft-law") to the formation of imperative norms within the framework of which the development and application of AI are subject to international legal norms.

On May 15, 2025, Ukraine signed the Convention on AI (requires ratification by the Verkhovna Rada), thereby confirming the political will and intention to integrate the principles defined in this document, in particular respect for human dignity, transparency, non-discrimination, protection of privacy, reliability and security, when adopting national legislation and practical application of AI in the public sector [2]. In this context, the signing of the Convention on AI became an important step in the process of European integration of Ukraine, which determines the vectors of reforming the legal system taking into account new digital risks and challenges.

In view of the above, there is a need to conduct a comprehensive scientific study of issues related to the analysis of the provisions of the Convention on AI, outlining their significance for ensuring and protecting human rights, as well as identifying challenges and threats to the legal system of Ukraine in the context of digital transformation.

Such research acquires particular importance in the context of Article 3 of the Constitution of Ukraine, which proclaims a person, his life, health, honor and dignity as the highest social value, and the establishment and protection of human rights and freedoms as the main duty of the state. Since the Council of Europe Framework Convention on Artificial Intelligence introduces a risk-based approach to algorithmic systems, its implementation in national legislation and law enforcement practice should be considered as one of the tools for specifying and practical implementation of the principle of the rule of law (Article 8 of the Constitution of Ukraine) and ensuring that the digitalization of public administration will not lead to a narrowing of the content and scope of existing human rights and freedoms (Part Three of Article 22 of the Constitution of Ukraine).

Research Aim and Objectives. The purpose of this article is to conduct a systematic legal analysis of the Convention on AI in the context of ensuring and protecting human rights, as well as to determine its significance and key challenges for the legal system of Ukraine in the context of digital transformation, taking into account modern approaches to AI governance and the principle of legal certainty of algorithmic decisions.

To achieve this goal, the following tasks were set and solved in the article: 1) the legal nature, goals and fundamental principles of the Convention on AI were studied, in particular in relation to the principles of the rule of law, democracy and respect for human rights; 2) an analysis of the correlation of the provisions of the Convention on AI with the main international acts in the field of regulating the protection of human rights and ensuring the rule of law was conducted; 3) potential risks and challenges to ensuring human rights that arise in the process of developing and applying AI were identified, as well as mechanisms for their minimization, provided for by the provisions of the Convention on AI; 4) the significance of the Convention on AI for Ukraine was determined, in particular in terms of implementing its provisions into national legislation, adapting law enforcement practice and fulfilling the state's European integration obligations; 5) proposals have been developed for further harmonization of national legislation with the provisions of the Convention on AI in order to strengthen human rights guarantees and ensure the rule of law in the context of the development of digital technologies and their use in public administration and other areas.

State of Research. Certain issues related to ensuring the protection of human rights when using AI, including the international legal aspect of this issue, have been studied in their works by such scholars as A. Bakumenko, A. Gachkevych, D. Honcharuk, O. Humin, N. Zagrebelna, N. Mazaraki, V. Olyukha, and others. Some aspects of this issue are also covered in the works of the author of this article, in particular, regarding the impact of AI on the constitutional and legal regulation of justice and the protection of human and civil rights [3], on the implementation of fundamental constitutional principles, in particular democracy, separation of powers, the rule of law, and legal certainty [4] , as well as on the system of human rights enshrined in the European Convention on Human Rights, in terms of risks to such rights as the right to respect for private life, freedom of expression, the prohibition of discrimination, and the right to an effective remedy [5] .

A review of existing scientific works in this area allows us to conclude that most scientists agree that the rapid development of AI and its integration into various areas of legal relations necessitates the urgent need to revise and adapt the national legislation of Ukraine.

In particular, some scholars (for example, A. V. Bakumenko, N. A. Zagrebelna, etc.) draw attention to the fact that Ukraine’s signing of the Convention on AI has become an imperative for conducting systematic monitoring of the provisions of the Criminal Procedure Code of Ukraine, the Law of Ukraine “On Operational and Investigative Activities”, as well as legislation on the protection of personal data in order to identify gaps in the regulation of digital evidence and the use of algorithmic systems. The authors emphasize the need to standardize the procedures for admissibility, certification and appeal of results obtained using AI, which is critically important for compliance with the principle of the rule of law and ensuring effective judicial protection [6, p. 14].

Similarly, in other scientific works (for example, by scientists such as O. M. Gumin, A. AT. Gachkevych et al.) outlines the issue of using AI in criminal justice, including predictive analytics, smart technologies and risk assessment systems. Scientists emphasize that technological progress can significantly expand the capabilities of law enforcement agencies, but at the same time creates risks of violating human rights and the rule of law. An important aspect of these studies is the systematization of international standards, in particular the provisions of the Convention on AI, which, according to scientists, should serve as a guideline for the implementation of modern technologies in criminal justice [7, p. 157].

Other scientists, in particular V. G. Olyukha, emphasize the supporting role of AI in the legislative process, in particular in the preparation of draft laws. Among the main functions of AI, scientists highlight such as analyzing the consistency of legislative proposals with current legislation and acts of the European Union (EU), identifying legislative gaps and potential conflicts, generating preliminary versions of texts for further processing, as well as increasing the efficiency of regulating digital readiness and compatibility with international standards. The growing role of AI in legislative work indicates the prospects for using algorithmic systems in the process of forming legal norms, while simultaneously outlining new risks regarding the accuracy and legal validity of AI recommendations [8, p. 218].

Special attention in scientific research is paid to the use of large language models (LLM) and generative AI in judicial practice. In particular, N. Mazaraki and D. Goncharuk emphasize that language models do not operate on legal knowledge as such, but only reveal statistical patterns in training data, which creates potential risks of bias, especially in cases of application of such language models in court cases, which, in turn, necessitates the formalization of control procedures and verification of the results obtained with the help of such models [9, p. 6].

At the same time, the issues of judicial review of algorithmic opacity (lack of transparency and explainability of algorithmic models) and the influence of the cognitive effect of automation bias on the process of making government and judicial decisions remain insufficiently developed in the scientific literature.

The research of such scientists as A.V. Kozhina, V.V. Orishchuk, S.M. Markina is devoted to digital transformation in the context of the implementation of AI and the Smart City concept in the EU. The authors analyze the practice of integrating modern information technologies, the Internet of Things and AI into city management, energy and transport, emphasizing the role of European platforms and programs (Horizon 2020, Horizon Europe, Living-in.EU) in ensuring standardized and effective use of technologies [10, p. 6].

However, despite a significant amount of scientific research in the field of legal regulation of the use of AI in public administration and in other areas, there remain a number of insufficiently studied problematic issues in the context of Ukraine's signing of the Convention on AI. In particular, in the science of constitutional law, the following issues remain poorly studied: determining ways to implement the provisions of the Convention on AI in the legislation of Ukraine; the correlation and consistency of the provisions of European standards in the field of human rights protection with the norms of the Convention on AI; legal assessment of risks and challenges to ensuring human rights that arise in the process of developing and applying AI, as well as mechanisms for their minimization, provided for by the provisions of the Convention on AI, etc.

Thus, a review of scientific developments indicates the objective need for a systematic legal study of the Council of Europe Framework Convention on Artificial Intelligence through the prism of its implementation in the legal system of Ukraine. Such a study involves a comprehensive analysis of not only the regulatory framework, but also the procedural and institutional mechanisms for the application of artificial intelligence technologies in order to ensure effective protection of human rights.

Presentation of the main material. The Convention on AI, by its legal nature, is an international multilateral treaty aimed at forming universal legally binding standards regulating the development, implementation and use of AI in public administration and other areas of public relations [11].

An analysis of the provisions of this Convention shows that it is aimed at establishing general legal standards and, at the same time, leaves room for states to choose specific mechanisms for implementing its principles into national legislation. In particular, the Convention on AI consists of a Preamble and eight chapters, each of which is devoted to a separate aspect of regulating the use of AI [11].

Thus, the Preamble enshrines the fundamental values on which the document is based, namely respect for human rights, democracy and the rule of law, references to key universal and European human rights treaties, which confirms the systemic connection of the Convention with the current international legal order, and the Preamble also outlines the need to respond to challenges related to the development and implementation of AI.

Section I, “General Provisions”, defines the subject matter and purpose of the Convention, as well as its scope. This section sets out the main definitions, in particular the concept of “artificial intelligence system”, which means a machine system that, with explicit or implicit purposes, determines, based on input data, what it receives, how to generate output data, in particular predictions, content, recommendations or decisions that can affect the physical or virtual environment (Art. 2 of the AI Convention) [11]

Section II, “General Obligations,” sets out a set of positive obligations of states to prevent human rights violations in the development, implementation, and use of AI. Particular attention is paid to risk assessment, the establishment of legal protection mechanisms, and the need to ensure effective means of appealing algorithmic decisions

A separate section contains a list of fundamental principles that should determine state policy and legal regulation in the field of AI use. The basic principles enshrined in the Convention on AI include, in particular, the principle of human-centricity, according to which AI technologies should be developed and applied taking into account the priority of human rights and freedoms, the principle of non-discrimination and equality, transparency and explainability of algorithmic decisions, accountability of entities using AI, as well as ensuring effective legal remedies. The set of these principles forms a normative model of responsible use of AI, focused on preventing systemic violations of human rights.

In this context, the doctrinal interpretation of the provisions of the AI Convention on human dignity, individual autonomy, accountability and procedural guarantees is of key importance as they implicitly require the preservation of the decisive role of humans in the decision-making process that may significantly affect the exercise of individual rights and freedoms. Although the Convention does not explicitly use the term human oversight, its provisions form the legal basis for preventing the complete autonomy of government decisions and ensuring the possibility of human intervention and control.

The AI Convention also defines mechanisms for cooperation between states, in particular, it establishes the basis for the exchange of information, best practices and the development of international coordination in the field of AI regulation, the necessity of which is due to the cross-border nature of digital technologies. This section also establishes procedural guarantees for individuals in cases where AI significantly affects the exercise of their rights.

Of particular importance is Chapter V, “Assessment and Mitigation of Risks and Adverse Impacts,” which obliges States to take or continue to take measures to identify, assess, prevent, and mitigate the risks posed by AI systems, taking into account the actual and potential impact on human rights, democracy, and the rule of law, including ongoing monitoring of risks and adverse impacts on these values.

A separate section is devoted to the procedure for implementing the provisions of the Convention, in particular, it is determined that each party, in accordance with its domestic legislation and applicable international obligations, shall duly take into account any specific needs and vulnerabilities in connection with the observance of the rights of persons with disabilities and children.

The AI Convention also contains two more chapters – Chapter VII “Follow-up Mechanism and Cooperation”, which defines the reporting obligations of the Parties, international cooperation and the establishment of independent national oversight mechanisms to ensure the effective implementation and compliance of States' obligations in the field of AI use, and Chapter VIII “Final Provisions”, which contains standard provisions for international law on the regulation of issues of signature, ratification, entry into force, amendment and denunciation of this Convention.

It is noteworthy that the process of developing the Convention on AI lasted five years and began in 2019, when, on behalf of the Committee of Ministers of the Council of Europe, the Ad Hoc Committee on Artificial Intelligence (CAHAI) was established [12], the main task of which was to study the feasibility of creating a legal instrument that would be based on existing EU standards in the field of human rights, democracy and the rule of law. In 2022, the work was continued by the Committee on Artificial Intelligence (CAI), which directly developed the text of the Convention on AI. It is important to emphasize that not only European countries, but also other countries of the world were involved in the process, which emphasizes the global nature of the instrument [13].

It is significant that the aim of the AI Convention was not to limit technological progress as such, but to direct it in a legal direction compatible with the values of democracy, the rule of law and respect for human dignity [12]. At the same time, given that the AI Convention is a framework, it establishes only the general, most fundamental principles and objectives, as well as methods of their implementation, and the specific legal and procedural mechanisms for their achievement are left to the discretion of the parties, taking into account the specifics of their national systems [11].

It should be noted that during the development of the provisions of the Convention on AI, attention was repeatedly drawn to the fact that the use of AI creates specific risks, in particular digital dehumanization, interference with privacy through unauthorized data processing, manipulation of electoral processes through deepfakes technologies, and restrictions on access to justice through the opacity of algorithmic decisions. In view of this, the Convention attempts to introduce a set of mechanisms to minimize and prevent these threats.

The implementation of the obligation of States to assess and mitigate risks, as stipulated in Article 16 of the Convention on AI, can in practice be carried out using specialized tools developed within the institutions of the Council of Europe. One such tool is the HUDERIA (Human Rights, Democracy and the Rule of Law Impact Assessment) methodology [15], prepared by the Committee on Artificial Intelligence (CAI) as an auxiliary, recommendatory mechanism for systematizing risk assessment, rather than as a legally binding element of the Convention itself.

The specified methodology provides for four stages of such assessment, namely:

1. Contextual Risk Analysis (COBRA), which identifies risk factors based on the context of application (e.g., healthcare, labor law, justice, etc.) and the development of AI technologies.

2. Stakeholder engagement (SEP) and mandatory consultations with groups of stakeholders whose rights may be affected by the implementation or use of AI.

3. Impact Assessment (RIA), i.e. conducting a systematic and detailed analysis of the scale, probability, and reversibility of potential harm from the implementation and use of AI.

4. Mitigation Plan (MP) by developing specific steps to minimize risks, including the possibility of establishing a moratorium or ban on the use of certain AI technologies that are recognized as incompatible with democratic values [15].

In addition to preventive measures aimed at assessing the risks to human rights when using AI, the AI Convention introduces procedural guarantees, in particular, such as the right of an individual to know that he or she is interacting with AI, the right to an effective remedy when using AI, including the possibility of appealing automated decisions to competent independent bodies, and guaranteeing the right to compensation for damage [11].

In the context of this study, it is also necessary to emphasize that the provisions of the Convention on AI were taken into account in the development of the provisions of already existing international instruments, some of which are referenced in the Preamble of this Convention. In view of this, it is necessary to determine the consistency of the Convention on AI with other international instruments in the field of human rights protection and the establishment of the rule of law.

First of all, it should be noted that the basis of the connection of this Convention with other international instruments is a common value base, based, in particular, on respect for human rights, democracy and the rule of law.

One of the first EU acts in this area was the EU Regulation on Artificial Intelligence of 2024 (AI Act), which established the fundamental rules for the functioning of the EU’s internal digital space. The AI Convention, unlike this Regulation, has a framework nature, defining general legal guidelines and minimum standards for the implementation and use of AI in public administration and in other areas. Meanwhile, the AI Convention does not repeat the provisions of the EU Regulation on AI, instead, its provisions contain general human rights provisions within the framework of which EU law should apply. The provisions of the AI Convention focus on the legal consequences of the use of AI and relate primarily to the protection of human rights and mechanisms of legal liability in the field of implementation and use of AI [16].

An important aspect of the relationship between the AI Convention and the EU AI Regulation is the issue of legal binding force, as EU regulations have direct effect in the legal systems of the Member States, while the Convention requires ratification and implementation. At the same time, its provisions create international legal obligations for states that are both EU members and states that have acceded to the Convention, which necessitates the consistent interpretation and application of EU acts with the provisions of the Convention.

In the strategic dimension, it is necessary to point out the coherence of the AI Convention with EU strategies in the field of digital development. This concerns, in particular, the European Digital Strategy and the Human-Centric AI Policy [17], which demonstrates the consistency of EU institutions, in particular the Committee of Ministers of the Council of Europe, in the field of AI regulation in terms of its impact on human rights, democracy and the rule of law.

At the same time, it should be noted that the Convention on AI operates in close regulatory connection with universal and regional international legal instruments in the field of human rights protection, taking into account the main provisions of such instruments and specifying already existing democratic standards in the context of digital transformation and the use of AI. The need for such an approach is due to the emergence of new forms of interference in the field of human rights that were not directly provided for by existing international instruments.

Thus, the Universal Declaration of Human Rights of 1948 [18], which constitutes the axiological basis of the entire modern system of international human rights protection, has a direct impact on the approach applied in the Convention on AI regarding the priority of such values as human dignity, equality and non-discrimination in all areas of legal relations, including the sphere of use of AI. The Convention on AI develops these provisions by establishing the obligation of states to prevent human rights violations that may arise as a result of decision-making in public administration based on AI, thus performing the function of a modern interpretation of the universal values of the Declaration in the digital age.

The AI Convention also takes into account the fundamental principles enshrined in the 1950 Convention for the Protection of Human Rights and Fundamental Freedoms (ETS No. 5) [19], which occupies a central place in the system of European legal protection of human rights. The AI Convention, by placing at the basis of legal regulation such principles as the right to respect for private life, the right to a fair trial and the prohibition of discrimination, clarifies the scope of positive obligations of states in the context of the application of AI.

The International Covenant on Civil and Political Rights of 1966 [20] also has a normative interaction with the AI Convention, as the Covenant’s provisions on the right to an effective remedy, equality before the law and freedom of expression take on new meaning in the context of the use of AI. In this regard, the provisions of the AI Convention aim to ensure the effective implementation of these rights in the digital environment, emphasizing the need for human control over algorithmic processes, which directly follows from the requirements of the Covenant.

The International Covenant on Economic, Social and Cultural Rights of 1966 defines the obligations of states to progressively implement social rights [21], thanks to which the provisions of the Convention on AI regulate the principles of human rights protection, in particular in the part concerning the use of AI in the fields of health care, education, social security and labor, which is due to the increased risk of increasing social inequality and digital discrimination.

The European Social Charter of 1961 (ETS No. 35) [22] and the European Social Charter (revised) of 1996 (ETS No. 163) [23] were of particular importance in shaping the provisions of the AI Convention, in particular, this concerns the use of AI in the field of employment, social protection and access to services, which may affect the implementation of social rights. Since the AI Convention emphasizes the need to ensure non-discriminatory access to such services and transparency of algorithmic decisions, which is fully consistent with the obligations of states under the European Social Charter.

Furthermore, the AI Convention has a close normative relationship with the 1981 Convention for the Protection of Individuals with regard to Automatic Processing of Personal Data (ETS No. 108) [24] and its protocols, as both instruments aim to ensure the protection of fundamental human rights in the context of digital information processing. ETS No. 108 establishes basic standards of lawfulness, fairness and proportionality for the processing of personal data, while the AI Convention extends these standards to a wider range of algorithmic processes that are not limited to data processing but include automated decision-making and prediction. The protocols to ETS No. 108, in particular the Additional Protocol to the Convention for the Protection of Individuals with regard to Automatic Processing of Personal Data concerning supervisory authorities and cross-border data flows [25], strengthen the requirements for transparency, accountability and oversight, and also expand the area of application of such requirements, which directly corresponds to the principles of explainability and human control over AI algorithms enshrined in the Convention on AI.

The relationship between the AI Convention and the 1989 UN Convention on the Rights of the Child [26] is of particular importance, as children are a particularly vulnerable category of persons in the digital environment, and therefore the use of AI in education, social services and digital platforms creates increased risks to their rights. Therefore, the provisions of the AI Convention are consistent with the principle of the best interests of the child and emphasize the need for special safeguards in the case of the use of algorithmic systems in relation to minors.

The 2006 UN Convention on the Rights of Persons with Disabilities [27] was also important in shaping the provisions of the AI Convention, as AI can both promote inclusion and create new barriers. In this regard, the AI Convention takes into account the provisions of the UN Convention on the Rights of Persons with Disabilities in terms of ensuring accessibility, equality and non-discrimination, as well as the need to take into account the needs of persons with disabilities in the development and application of AI.

Therefore, the provisions of the Convention on AI are characterized by systematic interaction and consistency with international, including European, acts in the field of human rights protection, taking into account the relevant provisions and adapting them specifically to the relationships that arise in the process of implementing and using AI.

However, despite the comprehensive and systematic approach taken in the Convention on AI to prevent and address the negative impacts on human rights, democracy and the rule of law from the use of AI in public administration and other areas, and to ensure the protection of human rights in the process of such use, potential risks and challenges to ensuring human rights in the implementation and use of AI still remain.

In this regard, it is important to recognize that the use of AI generates a complex of systemic risks for the implementation of fundamental human rights and freedoms, which cannot be adequately prevented solely by means of national law. One of the key challenges is the limitation of the principle of access to justice and effective judicial protection due to the opacity of algorithmic models, which makes it impossible to fully control the legality of automated decisions [1].

In addition, algorithmic discrimination poses a particular threat to the implementation of the principle of equality and the prohibition of discrimination, since the use of statistical models can lead to indirect restrictions on the rights of individuals or groups of individuals on grounds that are not formally taken into account, but actually affect the result, which creates the risk of systemic reproduction of social inequality in law enforcement practice [13].

In addition, the use of AI in the field of uncontrolled data collection and analysis entails increased risks of violations of the right to respect for the private life of an individual, in particular through automated profiling and prediction of the behavior of such individuals. In combination with the use of AI in the field of public administration, such practices may lead to the deformation of democratic processes and the undermining of trust in state institutions, which necessitates the introduction of special legal restrictions and control mechanisms enshrined in the Convention on AI.

Although the Convention on AI cannot fully eliminate the risks to human rights, democracy and the rule of law, it does form a multi-layered system of legal mechanisms to prevent and minimize them. First of all, this concerns the obligation of states to implement a risk-based approach to regulating AI, which involves identifying, assessing and managing risks to human rights throughout the lifecycle of such processes. The Convention on AI also establishes the need to ensure procedural accountability and the possibility of human intervention in the process of automated decision-making, which creates the legal prerequisites for judicial and administrative control.

Of particular importance is the introduction of requirements to prevent discriminatory consequences, which should be implemented through preventive measures, regular review of algorithmic systems and the availability of effective legal remedies. In addition, the Convention on AI provides for the creation of independent national control mechanisms and the development of international cooperation, which allows responding to cross-border risks and ensuring the consistent application of standards in the field of AI use [1].

Ukraine, which signed and plans to ratify the Convention on AI on May 15, 2025 [2; 28], has actually undertaken to implement the provisions of the Convention into its national legislation and, accordingly, to systematically review such legislation in order to integrate human rights protection standards in the field of digital technologies, in particular by establishing procedures for assessing the impact of AI on human rights and mechanisms of legal liability, as well as determining other guarantees for the protection of human rights in the process of implementing and using AI, in particular in public administration.

When characterizing the significance of the AI Convention for Ukraine, it is necessary to proceed from three main levers of its influence. First of all, it should be noted that the AI Convention has necessitated the adaptation of Ukraine's national legislation, in particular by developing a special law on AI by the end of 2026, which will be harmonized with both the AI Convention and the EU Regulation on Artificial Intelligence of 2024 (AI Act).

In the law enforcement aspect, the provisions of the AI Convention also form a methodological basis for the development of judicial practice regarding the control of automated decisions of public authorities (administrative bodies), the definition of standards of proof and the limits of permissible automation of the exercise of public authorities on the basis of the rule of law, and the provision of effective judicial protection.

In addition, in the context of Ukraine's European integration, the Convention on AI acts as an element of the fulfillment of the state's international obligations and confirms the orientation towards the implementation of European legal standards in the field of digital development. The implementation of its provisions creates the prerequisites for the formation of a national doctrine of legal regulation of AI, compatible with the European legal space and the practice of international judicial institutions. Today, these strategies are defined by the Ministry of Digital Transformation of Ukraine in the White Paper on the Regulation of AI in Ukraine, which provides for the preparation of business through voluntary standards and impact assessment methodologies for the introduction of mandatory norms in the field of regulating the use of AI [14].

In addition, despite Ukraine's signing of the AI Convention, the possibility was preserved for Ukraine to achieve a balance between innovative development and ensuring national security, since the Convention contains a reservation according to which its action does not extend to issues of national defense, which is especially important in conditions of martial law. At the same time, Ukraine declared its intention to extend the principles of the AI Convention to the private sector through appropriate measures that will stimulate business to adhere to ethical standards. Participation in the Convention strengthens Ukraine's authority as a technological state that shares democratic values and participates in shaping the global agenda for regulating future technologies.

At the same time, even the Convention’s national security and defense exception is not absolute, as Article 3 of the Convention requires that such activities be carried out in a manner compatible with international human rights law and with respect for democratic institutions. This provision significantly limits the possibility of abuse of the exceptions and preserves the potential for judicial and international oversight of the use of AI in the security sector.

Further implementation of the AI Convention, as the analysis conducted in the article shows, requires a systematic review of the national legislation of Ukraine and the adoption of the following measures:

1) formation of a holistic regulatory framework for regulating AI, in particular by adopting a special law on AI, which should enshrine the fundamental principles of the use of AI defined in the Convention on AI, in particular the principle of human-centricity, legal certainty, proportionality and accountability of AI algorithms to humans, as well as establish the limits of the permissible use of AI in public administration and identify areas of increased risk;

2) integration of human rights guarantees into some acts of current legislation, in particular those regulating administrative procedures, criminal proceedings, operational and investigative activities, justice, protection of personal data, etc. Such changes should provide for special procedural guarantees in the case of the use of AI, namely the right of a person to be informed about the use of algorithmic decisions, the right to an effective remedy, which includes the possibility of appealing decisions made or significantly determined by the use of artificial intelligence systems, and the provision of procedural guarantees that allow for their substantive verification by a person;

3) introduction of mechanisms for assessing the impact of AI on human rights by means of regulatory consolidation of the obligation of such an assessment before implementing AI in the activities of public authorities, which should be carried out taking into account the risks of discrimination, violation of privacy and restriction of access to justice, and its results should have legal significance for decision-making by public authorities in the relevant field;

4) strengthening institutional control and accountability over the use of AI in public administration, in particular by identifying a special body authorized to supervise the use of AI by public authorities, whose competence should include monitoring compliance with the standards of the Convention on AI (and in the future, the provisions of a special law on AI), considering complaints and preparing recommendations for relevant entities; such a specially authorized body will be required to operate in close cooperation with the Commissioner for Human Rights of the Verkhovna Rada of Ukraine;

5) adaptation of judicial and administrative practice to the realities of introducing AI into the sphere of public administration in order to form unified approaches to the assessment of evidence and decisions obtained using AI; development and improvement of existing recommendations on the application of the principles of the Convention on AI in law enforcement activities;

6) ensuring transparency and explainability of decisions made with the help of AI by imposing on government officials the obligation to ensure transparency in the use of AI; determining the permissible limits of the use of AI for decision-making by government officials and the possibility of their verification;

7) improving the legal culture and professional competence of persons authorized to perform state and local government functions by introducing systematic training of such persons on issues of legal regulation of AI, etc.

Summarizing the conducted research, it can be concluded that the harmonization of Ukraine’s national legislation with the provisions of the Convention on AI is of a comprehensive nature, aimed at ensuring effective protection of human rights and establishing the rule of law in the context of the digital transformation of public administration and in other areas, which corresponds to the European vector of Ukraine's development and its international legal obligations.

Conclusions

1. The Convention on AI captures the transition from declarative regulation of digital technologies to a model of legally binding, risk-based governance of algorithmic systems. This document is comprehensive in nature, establishing mandatory minimum standards for regulating activities related to the development, implementation and use of AI, regardless of their field of application, in order to ensure the compatibility of technological progress with the principles of the rule of law, democracy and the protection of human rights. The Convention on AI establishes a system of fundamental principles and procedural guarantees, among which the principle of respect for human dignity, the principle of legality, the principle of proportionality and the principle of non-discrimination, as well as the principles of transparency and explainability of algorithmic processes, which are aimed at ensuring the possibility of control over automated decisions of public authorities (administrative bodies). The AI Convention implicitly establishes a requirement for meaningful human oversight, which stems from its provisions on accountability, transparency and procedural safeguards. This implies preserving the decisive role of humans in making decisions that can significantly affect the exercise of rights and freedoms, thus creating a functional equivalent of the principle of “human oversight”. Among the key guarantees for the protection of human rights during the implementation and use of AI, the Convention enshrines the obligation of states to implement an assessment of the impact of AI technologies on human rights, democracy and the rule of law, as well as to ensure effective mechanisms for oversight, accountability and legal protection.

2. The AI Convention acts as a “bridge” between classic human rights protection instruments and new technological realities, ensuring the evolutionary renewal of the system of such protection. The provisions of this document take into account the fundamental approaches to the protection of human rights and the establishment of democratic values, enshrined in other international acts in the field of human rights. The Convention does not modify the content of fundamental rights and freedoms enshrined in the Universal Declaration of Human Rights, the European Convention on Human Rights, the UN International Covenants and the European Social Charter, etc., but ensures their adaptation to the conditions of algorithmic management and automated decision-making, thus acting as a special act of implementing general human rights standards in the field of digital technologies. In particular, the AI Convention specifies the content of the right to respect for human dignity, the right to privacy (in the broad sense of the inviolability of private life) and the protection of personal data, enshrined in the Universal Declaration of Human Rights, the European Convention on Human Rights and the International Covenant on Civil and Political Rights, in the context of automated data collection, analysis and use, and also adapts the principle of equality and non-discrimination, transforming it into requirements for preventing algorithmic bias and discriminatory consequences of automated decisions; The AI Convention extends the guarantees of the right to an effective remedy and a fair trial to situations involving automated decision-making, which corresponds to the provisions of Articles 6 and 13 of the Convention for the Protection of Human Rights and Fundamental Freedoms, and also adapts the socio-economic rights enshrined in the International Covenant on Economic, Social and Cultural Rights and the European Social Charter, taking into account the use of AI in the field of work, social security and access to administrative services.

3. AI impact assessment and human control requirements form a new standard of due process in automated decision-making. During the development of the AI Convention, it was recognized that AI technologies can systematically affect the realization of the right to effective judicial protection, the principle of equality, the prohibition of discrimination and the right to respect for private life, given that the Convention establishes a structured model for reducing these risks through a combination of preventive and corrective mechanisms, in particular, human rights impact assessment, requirements for transparency, human control, accountability and accessibility of appeal procedures. These mechanisms form the legal prerequisites for ensuring real, rather than formal, protection of the individual in the context of the use of algorithmic systems.

4. The Convention on AI is of significant importance for Ukraine in the context of reforming the national legal system and implementing international obligations in the field of human rights, as its provisions create a normative basis for revising sectoral legislation, in particular in the field of public administration, judicial proceedings, criminal justice and personal data protection. The Convention also contributes to the unification of approaches to the use of artificial intelligence in the activities of public authorities and the formation of uniform standards of law enforcement.

5. In order to further implement the provisions of the Convention on AI into national legislation, it is proposed to apply a comprehensive approach, which, in particular, includes: regulatory consolidation of general principles for the use of AI, in particular in legislation on administrative services and procedures, justice, etc.; legal consolidation of mandatory procedures for assessing the impact of AI on human rights, democracy and the rule of law, which should be applied to high-risk AI before their implementation and during operation; institutional strengthening of independent mechanisms for monitoring the use of AI, in particular by entrusting a special body with the authority to assess the impact of AI introduced into the activities of a government body and to monitor the use of such technologies; regulation of the use of AI in the activities of government entities, in particular by establishing requirements for transparency, explainability and the possibility of human intervention in the decision-making process; adaptation of procedural legislation to ensure effective means of judicial and administrative appeal of decisions made using algorithmic systems.

List of sources used:

Rotenberg M. Framework Convention on AI, Human Rights, Democracy and Rule of Law. International Legal Materials. 2025. Vol. 64, no. 3. P. 859–902. URL: https://www.cambridge.org/core/journals/international-legal-materials/article/framework-convention-on-artificial-intelligence-and-human-rights-democracy-and-the-rule-of-law-council-eur/0CCDA03299BF85537031F1CA26CF2CBD

Safe AI for millions of Ukrainians: Ukraine signed the Framework Convention on Artificial Intelligence and Human Rights: Government portal dated May 15, 2025. URL: https://www.kmu.gov.ua/news/bezpechnyi-shi-dlia-milioniv-ukraintsiv-ukraina-pidpysala-ramkovu-konventsiiu-pro-shtuchnyi-intelekt-ta-prava-liudyny

Bernaziuk I. M. Artificial intelligence in justice: constitutional and legal aspect. Analytical and comparative jurisprudence. 2025. No. 2. P. 130–137. URL: http://journal-app.uzhnu.edu.ua/article/view/326861

Bernaziuk I. M. Artificial intelligence as a new reality and its potential impact on constitutional and legal regulation in Ukraine. Electronic scientific publication "Analytical and comparative jurisprudence". 2025. No. 01. P. 76-82. URL: http://journal-app.uzhnu.edu.ua/article/view/323365

Bernaziuk I. M. Artificial intelligence and human rights: challenges for the European Convention on Human Rights // Analytical and comparative jurisprudence. 2025. Issue No. 03. P. 89-99. URL: http://journal-app.uzhnu.edu.ua/article/view/332656

Bakumenko A. V., Zagrebelna N. A. Artificial intelligence and legal challenges in the age of autonomous technologies. Law and society. 2025. No. 4. Vol. 1. P. 9-16. URL: http://pravoisuspilstvo.org.ua/archive/2025/4_2025/part_1/4.pdf

Gumin O. M., Gachkevych A. O. Application of modern technologies for the modernization of the criminal justice system and the standards of the Council of Europe on artificial intelligence. Bulletin of the Criminological Association of Ukraine. 2025. No. 2(35). Part 2. P. 147-161. URL: https://vca.univd.edu.ua/index.php/vca/article/view/595

Olyukha V. G. The use of artificial intelligence in lawmaking activities taking into account EU approaches. Scientific Bulletin of the Uzhhorod National University, 2025. Series LAW. Issue 91: part 1. P. 213-219. URL: http://visnyk-pravo.uzhnu.edu.ua/article/view/343479

Mazaraki N., Honcharuk D. Judicial AI and the irreparable bias problem // Access to Justice in Eastern Europe. 2025. Vol. 8, Spec. Issue. P. 1–21. URL: https://www.researchgate.net/publication/398737497_JUDICIAL_AI_AND_THE_IRREPARABLE_BIAS_PROBLEM

Kozhina A. V., Orishchuk V. V., Markina S. M. Analysis of the possibilities of using ethical artificial intelligence and Smart City tools for the restoration of Ukraine on the path to the EU: analytical study. Kyiv, IGS, 2025. 36 p. URL: https://www.researchgate.net/publication/397374198_Analiz_mozlivostej_zastosuvanna_eticnogo_stucnogo_intelektu_ta_instrumentiv_Smart_City_dla_vidnovlenna_Ukraini_na_slahu_do_ES

Council of Europe Framework Convention on Artificial Intelligence, Human Rights, Democracy and the Rule of Law , adopted at the 133rd meeting of the Committee of Ministers of the Council of Europe ( Strasbourg, 17 May 2024). CM(2024)52. URL: https://vkksu.gov.ua/sites/default/files/rye_ramkova_konvenciya_zi_shtuchnogo_intelektu_prav_lyudyny_demokratii_ta_verh._prava.pdf

Explanatory Report to the Council of Europe Framework Convention on Artificial Intelligence and Human Rights, Democracy and the Rule of Law. Council of Europe Treaty Series. No. 225. URL: https://rm.coe.int/1680afae67

The Framework Convention on Artificial Intelligence. Artificial Intelligence. URL: https://www.coe.int/en/web/artificial-intelligence/the-framework-convention-on-artificial-intelligence

White Paper on the Regulation of Artificial Intelligence in Ukraine. Ministry of Digital Transformation of Ukraine. 2024. URL: https://thedigital.gov.ua/news/technologies/regulyuvannya-shtuchnogo-intelektu-v-ukraini-prezentuemo-bilu-knigu

HUDERIA: Human Rights, Democracy and the Rule of Law Impact Assessment for AI Systems: methodology adopted by the Committee on Artificial Intelligence (CAI) of the Council of Europe. Council of Europe. 2024. URL: https://rm.coe.int/cai-2024-16rev2-methodology-for-the-risk-and-impact-assessment-of-arti/1680b2a09f

Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 laying down harmonized rules on artificial intelligence (Artificial Intelligence Act). Official Journal of the European Union. 2024. L series. URL: https://eur-lex.europa.eu/eli/reg/2024/1689/oj/eng

Shaping Europe's digital future: Communication from the European Commission of 19 February 2020. COM(2020)67final. EUR-Lex. 2020. URL https://eur-lex.europa.eu/legal-content/UK/TXT/?uri=CELEX:52020DC0067

Universal Declaration of Human Rights: adopted and proclaimed by resolution 217 A (III) of the UN General Assembly of December 10, 1948. URL: https://zakon.rada.gov.ua/laws/show/995_015#Text

Convention for the Protection of Human Rights and Fundamental Freedoms: international treaty of the Council of Europe of 4 November 1950 (ETS No. 5). URL: https://zakon.rada.gov.ua/laws/show/995_004#Text

International Covenant on Civil and Political Rights: adopted by the UN General Assembly on December 16, 1966. URL: https://zakon.rada.gov.ua/laws/show/995_043#Text

International Covenant on Economic, Social and Cultural Rights: adopted by the UN General Assembly on December 16, 1966. URL: https://zakon.rada.gov.ua/laws/show/995_042#Text

Social Charter : international treaty of the Council of Europe of 18 October 1961 (ETS No. 35). URL: https://zakon.rada.gov.ua/laws/show/994_300#Text

European Social Charter (revised): international treaty of the Council of Europe of 3 May 1996 (ETS No. 163). URL: https://zakon.rada.gov.ua/laws/show/994_062#Text

Convention for the Protection of Individuals with regard to Automatic Processing of Personal Data: signed by the Council of Europe on 28 January 1981. URL: https://zakon.rada.gov.ua/laws/show/994_326#Text .

Additional Protocol to the Convention for the Protection of Individuals with regard to Automatic Processing of Personal Data concerning supervisory authorities and cross-border data flows: signed by the Council of Europe on November 8, 2001. URL: https://zakon.rada.gov.ua/laws/show/994_363#Text .

Convention on the Rights of the Child: adopted by the UN General Assembly on November 20, 1989. URL: https://zakon.rada.gov.ua/laws/show/995_021#Text .

Convention on the Rights of Persons with Disabilities: adopted by the UN General Assembly on December 13, 2006. URL: https://zakon.rada.gov.ua/laws/show/995_g71#Text .

O. Bornyakov to sign the Framework Convention of the Council of Europe on Artificial Intelligence and Human Rights, Democracy and the Rule of Law: Decree of the President of Ukraine dated 12.05.2025 No. 54 / 2025 - рп. URL: https://zakon.rada.gov.ua/laws/show/54/2025-%D1%80%D0%BF#Text.