Ukraine - AI Regulation White Paper

Overview

The White Paper on Artificial Intelligence Regulation in Ukraine was published on June 26, 2024, by the Ministry of Digital Transformation in collaboration with the Expert Committee on the Development of Artificial Intelligence, representing a comprehensive implementation document for the October 2023 Roadmap's first stage. This policy framework provides detailed specifications for voluntary preparatory tools operating during 2024-2025 before binding legislation takes effect, elaborating what Ministry officials describe as a "soft approach" or "bottom-up" regulatory philosophy. The White Paper emerged from extensive consultations with civil society organizations, business stakeholders, legal experts, and international partners, with development support provided by USAID's Cybersecurity for Critical Infrastructure in Ukraine Activity and the Digital Transformation Activity supported by USAID and UK Aid. The document articulates three primary objectives: supporting Ukrainian AI industry competitiveness through tools enabling international market access, protecting citizens' rights from AI-related risks through appropriate safeguards, and advancing European integration through future EU AI Act implementation. The White Paper provides detailed specifications for the regulatory sandbox enabling developers to test systems with medium or high human rights impact, labeling mechanisms promoting transparency, frameworks for voluntary industry codes of conduct, and integration of sectoral guidelines for media, marketing, and education. It estimates Stage One preparatory phase duration at one to two years, though acknowledges this timeline may adjust based on EU AI Act finalization, business readiness assessments, and Ukraine's unique circumstances confronting ongoing war with Russia while simultaneously developing AI governance. The White Paper emphasizes that Ukrainian AI regulation serves dual purposes: domestic rights protection and international market access facilitation, as companies targeting EU markets must comply with EU AI Act regardless of domestic law, making preparatory alignment economically imperative. Legal expert Hordii Rumiantsev noted the White Paper "aims to balance the interests of everyone involved," while business leaders raised concerns about categorization clarity that the Ministry acknowledged while maintaining the framework's core approach.

Definitions

The White Paper establishes terminology building on the Concept and Roadmap foundations. Stage One Preparatory Phase refers to the initial implementation period (2024-2025, potentially extending to 2026) when voluntary tools operate without legal mandate, preparing businesses and government for eventual binding requirements. Extra-Legislative Tools encompasses preparatory measures including regulatory sandbox, labeling mechanisms, voluntary codes, and sectoral guidelines provided before formal legislation enactment. Soft Approach describes Ukraine's regulatory philosophy emphasizing voluntary compliance and business support before introducing binding legal requirements, contrasting with immediate comprehensive regulation. Self-Regulation means industry-led governance where companies establish and enforce standards among themselves with government facilitation but without direct legal mandate. Co-Regulation involves collaborative standard-setting where government and industry work jointly to develop governance mechanisms, sharing responsibility for AI system oversight. Regulatory Sandbox constitutes a controlled testing environment where AI developers evaluate systems against future regulatory requirements, receiving expert feedback before full market deployment, with entry restricted to systems with medium or high human rights impact. AI System Labeling refers to mechanisms enabling consumers and users to identify AI-powered products and services, promoting transparency about automated decision-making and AI involvement in services they use. Voluntary Code of Conduct means industry-developed standards for responsible AI development and deployment adopted without legal requirement but encouraged through government recognition and support. Sectoral Guidelines are domain-specific recommendations for responsible AI use adapted to particular fields' unique characteristics and challenges, such as media's editorial independence concerns or education's developmental considerations. Medium Human Rights Impact and High Human Rights Impact categorize AI systems based on potential to affect fundamental rights, with definitions elaborated through assessment methodologies to determine regulatory sandbox eligibility. Bottom-Up Implementation describes regulatory development proceeding from business practices and industry self-regulation toward government-imposed binding requirements, allowing organic evolution of governance mechanisms appropriate to Ukrainian context.

Governance and Institutional Framework

The White Paper's governance structure centers on the Ministry of Digital Transformation as primary implementation coordinator, working closely with the Expert Committee on the Development of Artificial Intelligence in Ukraine and its four specialized working groups covering public administration, education and science, security and defense, and regulation. The Ministry serves what the White Paper describes as a "service function," providing tools, guidance, and support to businesses preparing for future regulation rather than immediately imposing binding requirements. The Expert Committee, chaired by Oleksiy Molchanovskyi with Timofiy Milovanov as Co-Chairman for Regulation, coordinates voluntary code development, assesses regulatory sandbox applications, and provides guidance on sectoral guideline implementation. Regulatory Sandbox Oversight involves Expert Committee evaluation of applications from AI developers seeking to test systems with medium or high human rights impact, providing structured feedback on compliance prospects under future legislation. Sectoral Regulator Engagement connects the framework to domain-specific agencies: the National Council of Ukraine on Television and Radio Broadcasting for media guidelines, the Ministry of Education and Science for educational AI applications, and various industry regulators for sector-specific implementations. Industry Association Participation brings specialized expertise through organizations like the IT Ukraine Association for technology sector perspectives, the Interactive Advertising Bureau of Ukraine for marketing applications, the All-Ukrainian Advertising Coalition for advertising standards, the Ukrainian Marketing Association for commercial use cases, and the Ukrainian Bar Association for legal compliance considerations. Civil Society Monitoring ensures human rights organizations and advocacy groups maintain oversight on rights protection aspects, providing accountability for the White Paper's commitments to fundamental rights safeguarding. International Partner Engagement involves USAID providing capacity-building support, UK Aid funding development activities, EU institutions offering regulatory harmonization guidance, and Council of Europe initiatives like HUDERIA providing assessment methodologies. The Cabinet of Ministers maintains strategic oversight and will approve Stage Two legislative proposals when the preparatory phase concludes. This multi-stakeholder governance model reflects the White Paper's philosophical commitment to collaborative standard-setting rather than top-down regulatory imposition, ensuring diverse perspectives inform implementation while maintaining governmental coordination and eventual authority for binding legislation.

Key Focus Areas

• Regulatory Sandbox Operationalization: represents a central focus, specifying that the sandbox will accept AI systems with medium or high human rights impact, providing developers structured evaluation against future regulatory requirements. The White Paper elaborates selection criteria emphasizing human rights implications rather than technical characteristics alone, ensuring the sandbox addresses systems with greatest potential societal impact.
• Assessment Methodology Development and Dissemination: involves creating and publishing frameworks enabling businesses to self-evaluate AI systems' risk levels, compliance prospects, and human rights impacts, empowering proactive preparation rather than reactive adaptation.
• Labeling Mechanism Implementation: focuses on establishing practical systems for identifying AI-powered products and services in ways consumers can easily understand, balancing transparency promotion with avoiding excessive complexity or user burden.
• Voluntary Code Facilitation: emphasizes government support for industry-developed standards, providing recognition, coordination platforms, and encouragement without legally mandating adoption, building self-regulation culture that will ease eventual transition to binding requirements.
• Sectoral Guideline Integration: connects general AI governance principles to domain-specific contexts, with the White Paper explaining how media guidelines address editorial independence and misinformation concerns, marketing guidelines address consumer protection and manipulation risks, and education guidelines address developmental appropriateness and equity considerations.
• EU AI Act Preparation: represents a critical focus area, as the White Paper provides extensive analysis of EU regulatory requirements, helping Ukrainian businesses understand coming European standards and prepare for compliance regardless of whether they operate in EU markets, given the Act's extraterritorial reach.
• Human Rights Impact Assessment: elaborates methodologies for evaluating AI systems' potential effects on privacy, non-discrimination, free expression, due process, and other fundamental rights, operationalizing the abstract rights commitments from earlier strategic documents.
• Business Capacity Building: involves education, training, and resource provision helping Ukrainian AI companies develop compliance capabilities, ethical frameworks, and responsible development practices during the voluntary preparatory period.
• Public Awareness and Trust Building: addresses citizen understanding of AI technologies, benefits, and risks, promoting informed engagement with AI-powered services and democratic participation in governance conversations.
• International Market Access Support: provides Ukrainian companies with knowledge, tools, and frameworks necessary to enter global markets, particularly EU markets requiring AI Act compliance, positioning regulatory alignment as competitive advantage rather than burden.
• Wartime AI Governance Considerations: acknowledges Ukraine's unique circumstances using AI for defense and wartime governance, noting that defense applications remain outside civilian regulatory frameworks while civilian AI use requires protection even during wartime.

Implementation Framework

The White Paper's implementation proceeds through structured phases with defined deliverables and timelines. Immediate Actions (Q3-Q4 2024) include regulatory sandbox launch accepting initial applications from AI developers with medium or high human rights impact systems, publication of assessment methodologies enabling business self-evaluation, deployment of labeling mechanism pilot programs in selected sectors, and continued refinement of sectoral guidelines based on stakeholder feedback. Medium-Term Actions (2025) involve expanding regulatory sandbox participation as capacity allows, promoting voluntary code adoption through industry engagement and government recognition programs, conducting comprehensive business readiness assessments to evaluate preparation for binding regulation, monitoring EU AI Act finalization and analyzing implications for Ukrainian legislative proposals, and beginning technical work on Stage Two draft legislation informed by Stage One implementation lessons. Transition Planning (Late 2025-Early 2026) encompasses final preparatory phase assessments determining whether businesses and government are ready for binding requirements, public consultations on draft Stage Two legislation incorporating voluntary phase experiences, Cabinet of Ministers review and approval of legislative proposals for Verkhovna Rada submission, and communication campaigns preparing stakeholders for transition from voluntary to mandatory compliance. Continuous Activities Throughout Stage One include stakeholder engagement through regular consultations with business, civil society, and international partners; progress monitoring tracking regulatory sandbox outcomes, voluntary code adoption rates, and rights protection effectiveness; international alignment monitoring EU AI Act developments and global regulatory trends; capacity building strengthening business and government capabilities; and adaptive management allowing timeline and approach adjustments based on implementation realities and external developments. Success Criteria for Stage One identified in the White Paper include significant business participation in voluntary tools demonstrating engagement with preparatory frameworks, demonstrated understanding of future regulatory requirements through business readiness assessments, established self-regulation culture with industry standards gaining meaningful adoption, effective human rights protection during voluntary phase showing frameworks' adequacy, maintained international competitiveness with Ukrainian AI companies successfully accessing global markets, and stakeholder consensus supporting transition to binding legislation when Stage Two begins. The implementation framework explicitly acknowledges Ukraine's wartime context may require timeline flexibility while maintaining core commitments to rights protection and eventual EU alignment, balancing regulatory development with national survival imperatives.

Monitoring and Evaluation

The White Paper establishes monitoring mechanisms tracking Stage One implementation effectiveness and informing transition to Stage Two. Regulatory Sandbox Monitoring tracks applications received, systems evaluated, feedback provided, and compliance improvements demonstrated by participating developers, assessing whether the sandbox effectively prepares businesses for future requirements. Voluntary Code Adoption Tracking measures industry uptake of self-regulatory standards through surveys, industry association reporting, and direct engagement, identifying sectors with high adoption and those requiring additional incentives or support. Labeling Mechanism Effectiveness Assessment evaluates consumer awareness and understanding of AI system identification, measuring whether transparency mechanisms meaningfully inform user choices and build trust. Sectoral Guideline Impact Evaluation examines how domain-specific recommendations affect AI deployment practices, rights protection, and sector-specific outcomes in media, marketing, education, and other fields covered by guidelines. Human Rights Impact Monitoring tracks AI systems' effects on privacy, non-discrimination, free expression, and other fundamental rights through civil society reporting, user complaints, and proactive assessments, identifying emerging concerns requiring regulatory attention. Business Readiness Assessments occur at regular intervals, evaluating Ukrainian AI companies' understanding of future requirements, capability to comply with binding regulation, and general preparedness for Stage Two transition. International Market Access Tracking monitors Ukrainian companies' success entering global markets, particularly EU markets requiring AI Act compliance, assessing whether regulatory alignment facilitates or hinders commercial objectives. Stakeholder Feedback Collection maintains continuous dialogue with businesses, civil society, international partners, and government agencies, gathering qualitative insights on implementation challenges, emerging issues, and necessary adjustments. EU AI Act Alignment Monitoring continuously tracks European regulatory developments, assessing implications for Ukrainian approach and updating guidance to businesses accordingly. Quarterly Progress Reports provide structured updates on implementation status, challenges encountered, adjustments made, and timeline prospects, maintaining transparency and enabling adaptive management. Annual Strategic Reviews involve comprehensive assessments presented to Cabinet of Ministers, evaluating whether Stage One is achieving objectives and providing recommendations on Stage Two transition timing and legislative content. Comparative Benchmarking examines other countries' AI regulatory experiences, particularly EU candidate countries and regional peers, identifying lessons applicable to Ukrainian context. Public Reporting and Transparency ensures regular publication of monitoring results through government websites, stakeholder briefings, and public consultations, building trust and democratic accountability. The monitoring framework recognizes Ukraine's wartime circumstances may affect data collection and stakeholder engagement capacity while maintaining commitment to evidence-based regulatory development.

Penalties, Liability, and Appeals

The White Paper maintains the Roadmap's approach that Stage One operates through incentives rather than penalties, with specific liability regimes and enforcement mechanisms reserved for Stage Two binding legislation. Stage One Non-Mandatory Status means participation in regulatory sandbox, voluntary code adoption, sectoral guideline compliance, and labeling mechanism use carry no legal requirement during the preparatory phase, with non-participation resulting in no penalties though potentially reducing government support access or limiting preparation for future mandatory requirements. Regulatory Sandbox as Learning Environment emphasizes the sandbox provides guidance and feedback rather than enforcement, with participation helping developers understand compliance prospects without imposing consequences for systems not meeting future standards during testing phase. Voluntary Code Enforcement Through Reputation relies on industry self-monitoring and peer accountability rather than government penalties, with adoption incentivized through public recognition, industry leadership positioning, and consumer trust rather than legal mandate. Future Liability Framework Principles outlined in the White Paper anticipate Stage Two legislation will establish comprehensive liability regimes addressing responsibility when AI systems cause harm through malfunction, bias, privacy violations, discrimination, or other negative impacts. Developer Accountability will include obligations for AI system designers covering appropriate safeguard integration, risk assessment conduct, transparency provision about capabilities and limitations, and documentation maintenance enabling accountability. Deployer Responsibilities will address organizations implementing AI systems, including training requirements, monitoring obligations, human oversight maintenance, and intervention capability in automated processes. Harm Redress Mechanisms will enable individuals affected by AI system decisions or actions to seek remedies, with particular emphasis on high-stakes domains like employment, social benefits, healthcare, criminal justice, and financial services. Appeal Processes will provide pathways for contesting automated decisions, with requirements for human review of consequential AI-driven outcomes and clear procedures for challenging system outputs. Regulatory Oversight and Enforcement will empower sector-specific agencies to investigate complaints, conduct audits, impose proportionate penalties for non-compliance, and order corrective actions when AI systems violate legal requirements. Proportionality Principles emphasize future penalties will be calibrated to violation severity, business size, whether non-compliance was intentional or negligent, cooperation with investigations, and corrective measures implemented. Transitional Protections commit to reasonable compliance timelines when Stage Two requirements take effect, avoiding sudden disruption and recognizing companies participating in voluntary Stage One programs. The White Paper emphasizes eventual liability frameworks must balance innovation encouragement with meaningful rights protection, learning from EU AI Act approaches while adapting to Ukrainian legal traditions and institutional capacities. International Liability Alignment ensures Ukrainian liability regimes will be compatible with EU and other international frameworks, facilitating cross-border commerce and preventing regulatory arbitrage.

Relationship to Other Instruments

The White Paper on Artificial Intelligence Regulation in Ukraine elaborates and operationalizes the Roadmap for the Regulation of Artificial Intelligence in Ukraine presented in October 2023, translating the Roadmap's strategic framework into detailed implementation specifications for Stage One preparatory phase. The White Paper builds on foundations established in the Concept of Artificial Intelligence Development in Ukraine approved in December 2020, demonstrating progression from strategic vision through implementation framework to operational guidance across Ukraine's AI governance evolution. The White Paper connects to sectoral guidelines including the Guidelines for the Responsible Use of Artificial Intelligence in Media and Guidelines for the Responsible Use of AI in Advertising and Marketing, explaining how domain-specific recommendations integrate with overarching regulatory principles and Stage One voluntary tools. The document explicitly prepares for Stage Two alignment with the EU Artificial Intelligence Act, providing extensive analysis of European requirements and helping Ukrainian businesses understand coming standards, positioning eventual legislative proposals to mirror EU frameworks. The White Paper relates to Draft Law No. 8153 on Protection of Personal Data, Ukraine's GDPR-aligned data protection legislation addressing AI-relevant issues of automated processing, algorithmic decision-making, and profiling, providing complementary privacy framework necessary for comprehensive AI governance. Ukraine's signature of the Council of Europe Framework Convention on AI, Human Rights, Democracy, and Rule of Law in May 2025 provides international treaty foundation for the White Paper's human rights protection commitments and democratic values emphasis. The White Paper incorporates HUDERIA (Human Rights, Democracy and Rule of Law Impact Assessment) methodologies developed by Council of Europe, operationalizing these tools within Ukraine's regulatory sandbox and assessment frameworks. The document aligns with OECD AI Principles emphasizing human-centered values, transparency, robustness, accountability, and fundamental rights respect, ensuring Ukrainian approaches reflect global consensus. The White Paper connects to UNESCO Recommendations on the Ethics of AI, particularly provisions addressing cultural diversity, inclusion, and human dignity in AI deployment. Relationship to Ukraine's Digital Transformation Strategy positions AI regulation as integral component of broader digitalization initiatives championed by the Ministry, supporting e-governance, digital infrastructure development, and IT sector growth. National Security Connections acknowledge AI's defense applications particularly prominent since Russia's 2022 invasion, though the White Paper maintains that defense-sector AI remains outside civilian regulatory frameworks while civilian applications require protection even during wartime. The White Paper's implementation through USAID and UK Aid support reflects international partnership frameworks underpinning Ukraine's governance development, connecting domestic regulatory initiatives to Western institutional support and capacity-building programs.

International Alignment

The White Paper prioritizes international regulatory alignment, particularly with European Union standards, as both strategic objective and economic necessity for Ukrainian AI sector. EU AI Act Detailed Analysis constitutes a substantial portion of the White Paper, providing comprehensive examination of European regulatory requirements, risk classification systems, conformity assessment procedures, and compliance obligations, enabling Ukrainian businesses to understand and prepare for standards applicable when operating in EU markets regardless of domestic Ukrainian law due to the Act's extraterritorial application. Regulatory Harmonization Strategy commits that Stage Two binding legislation will mirror EU AI Act provisions, ensuring Ukrainian regulatory framework is compatible with European standards, facilitating seamless market access for Ukrainian companies and supporting Ukraine's broader European integration objectives as EU candidate country. OECD AI Principles Integration ensures the White Paper's voluntary codes, assessment methodologies, and sectoral guidelines operationalize OECD commitments to human-centered AI, transparency, robustness, security, accountability, and fundamental rights respect, demonstrating alignment with global consensus standards. Council of Europe Framework Implementation connects the White Paper to Ukraine's May 2025 signature of the Framework Convention on AI, Human Rights, Democracy, and Rule of Law, the first international legally binding AI treaty, with the White Paper's human rights assessment methodologies and regulatory sandbox selection criteria operationalizing Convention commitments. HUDERIA Methodology Adoption incorporates Council of Europe's Human Rights, Democracy and Rule of Law Impact Assessment tools within Ukraine's regulatory sandbox and broader assessment frameworks, participating in international pilot initiatives while adapting methodologies to Ukrainian context. UNESCO Ethics Alignment addresses cultural diversity, inclusion, transparency, responsibility, and monitoring principles from UNESCO Recommendations on the Ethics of AI, particularly relevant for sectoral guidelines in media and education. International Technical Standards commits to adopting relevant ISO, IEC, and other consensus technical standards for AI systems, facilitating interoperability, reducing trade barriers, and enabling Ukrainian companies to participate in global supply chains and markets. Multi-lateral Governance Forum Participation positions Ukraine as active contributor to evolving international AI norms through OECD AI working groups, UNESCO AI ethics initiatives, Council of Europe standard-setting processes, and regional cooperation frameworks. Transatlantic Partnership extends beyond Europe through USAID support reflecting United States engagement with Ukraine's AI governance development, diversifying international alignment beyond purely European focus. Peer Learning from Regulatory Pioneers examines experiences from countries that have enacted or are developing AI regulation, including EU member states implementing the AI Act, Canada developing AI and Data Act, Singapore implementing AI governance frameworks, and regional peers in Eastern Europe pursuing similar alignment strategies. The White Paper's international orientation serves not only technical regulatory purposes but also broader geopolitical objectives of anchoring Ukraine within Euro-Atlantic institutional frameworks, democratic governance standards, and Western technological ecosystems, particularly significant given Russia's invasion and Ukraine's determination to demonstrate commitment to European values and standards despite wartime circumstances.

Implementation Timeline

Sources and References