AI Regulation And The Ministry Of Artificial Intelligence In Kazakhstan

Introduction

Kazakhstan is emerging as a prominent example of state-led digital transformation, where artificial intelligence (“AI”) has moved from a technological experiment to a strategic tool of governance and economic modernisation. With over 90 percent of public services available online through the national eGov platform and private applications ¹, the country has created a strong digital foundation for integrating AI into administrative processes and service delivery.

Globally, governments are increasingly investing in AI to transform public administration and streamline decision-making ². Research indicates that AI adoption in the public sector can significantly accelerate processes, reduce backlogs, and improve policy responsiveness ³. Yet, many administrations still face a persistent challenge - a skills and training gap that hinders the full and consistent realisation of AI's potential ⁴.

Kazakhstan stands out as one of the few countries actively addressing these challenges through comprehensive state coordination and policy reform. In 2025, the government adopted a national AI strategy and legal framework that embedded AI into the country's institutional architecture and regulatory system. That same year, it established the Ministry of Artificial Intelligence and Digital Development, tasked with implementing ethical and transparent AI use across the economy and public sector ⁵.

This article examines how Kazakhstan's evolving AI ecosystem - its institutional design, legal framework, national priorities, and regulatory approach - is shaping a new model of state-driven, responsible artificial-intelligence governance.

1. Institutional Framework: The Ministry of Artificial Intelligence and Digital Development

Kazakhstan's establishment of the Ministry of Artificial Intelligence and Digital Development (the “MAIDD”) under Presidential Decree No. 997 of 18 September 2025 marks a key milestone in its modernisation agenda. By consolidating digital governance, data policy, and AI regulation within a single institutional framework, the state introduced a centralised model rarely implemented at comparable scale.

1.1. Mandate and Strategic Role

The MAIDD serves as the central authority for AI and digital development policy, shaping the strategic, regulatory, and infrastructural pillars of transformation. Its mission is to ensure that AI supports economic diversification, improved governance, and enhanced quality of life.

MAIDD its work rests on three dimensions:

1. development and enforcement of AI legislation, including the Law on Artificial Intelligence (2025);
2. management of digital infrastructure, including the National AI Platform, data centers, and AI clusters such as Alem.Cloud and AI-Farabium;
3. human capital development through programs such as AI Movement and AI-Sana.

The Ministry acts as both policy designer and ecosystem coordinator, working with the Ministry of Science and Higher Education on research and standards, with National Information Technologies JSC (“NITEC”) on e-government infrastructure, and with Astana Hub on AI innovation and startup support.

1.2. Coordinated Ecosystem and Governance

To prevent fragmented governance, the MAIDD oversees an integrated digital ecosystem. The Center for Digital Government Development pilots AI services and supports institutional transformation, while NITEC operates the National AI Platform and manages public datasets and secure data exchange. Strategic alignment is reinforced through coordination with the Presidential Council for AI Development.

1.3. Ensuring Security, Ethics, and Compliance

The Ministry's structure includes oversight mechanisms to safeguard digital sovereignty and cybersecurity. The Committee for Information Security protects national digital infrastructure and state information systems.

Under the AI Law's risk-based model, AI systems are classified as minimal, medium, or high risk. High-risk systems are subject to sectoral audits and inclusion in monitored lists, requiring assessment of data-library quality, statutory compliance, and the absence of prohibited functionalities. Oversight intensity therefore increases proportionately with system autonomy and potential societal impact.

2. The Legal Framework: Law “On Artificial Intelligence” and the Digital Code

Kazakhstan's digital governance framework entered a new phase in 2025 with the adoption of two cornerstone instruments: the Law “On Artificial Intelligence” (No. 230-VIII ZRK) and the Digital Code of the Republic of Kazakhstan, effective from July 2026. Together, they establish an integrated legal foundation for the responsible use of AI and digital technologies, balancing innovation with ethics, accountability, and citizen protection.

2.1. The AI Law: Principles and Regulatory Design

Signed on 17 November 2025 and effective from 18 January 2026, the Law “On Artificial Intelligence” is the first comprehensive AI statute in Central Asia. It defines the rights and obligations of developers, owners, and users, and sets lifecycle requirements for safety, reliability, transparency, and lawful operation.

The law is grounded in seven principles: legality; fairness and equality; transparency and explainability; accountability; protection of human welfare and autonomy; data confidentiality; and security. Article 7 requires users to receive clear information about an AI system's functions and limitations prior to use, ensuring informed interaction.

System owners and operators must maintain documentation, conduct annual risk assessments, and implement continuous monitoring. Article 18 defines risk management as an ongoing process of identifying, evaluating, mitigating, and reviewing AI-related risks.

2.2. Risk-Based Oversight and High-Risk Systems

A central feature of the law is its risk-based model. AI systems are classified as minimal, medium, or high risk depending on their potential impact on safety, societal welfare, or national security.

• Minimal risk: negligible social or economic impact.
• Medium risk: potential moderate harm or material damage.
• High risk: threats to safety, critical infrastructure, or national interests.

Articles 17–20 establish enhanced supervision for high-risk systems, including sectoral audits, continuous monitoring, and possible inclusion in sectoral lists of trusted high-risk AI systems maintained by state bodies. Inclusion requires assessment of data-library quality, algorithmic integrity, and the absence of prohibited functionalities, ensuring proportional oversight.

Article 17(3) prohibits manipulative AI practices, unauthorised biometric processing, social scoring, and emotion recognition without consent. Article 25 establishes the National Artificial Intelligence Platform as a controlled environment for the development, training, and verification of AI systems prior to deployment.

2.3. Integration with the Digital Code

The Digital Code, adopted in January 2026, complements the AI Law by providing a unified framework for digital rights, data governance, and algorithmic decision-making. It codifies principles such as technological neutrality, human-rights protection, data freedom, and proportional state participation.

The Code regulates algorithmic and automated decision-making affecting individuals. Persons are entitled to:

• notification when algorithmic systems are applied;
• request human review;
• receive an explanation of the main influencing factors, subject to confidentiality limits.

The Code also formalises digital identity and authentication, recognising digital confirmation mechanisms and smart contracts, provided they include human dispute-resolution safeguards. It confirms the legal equivalence of accredited electronic digital signatures to handwritten signatures.

Conclusion

Kazakhstan has moved from digital service delivery to institutionalised AI governance, embedding artificial intelligence within its core regulatory and administrative structures.

Through the Ministry of Artificial Intelligence and Digital Development, the AI Law, and the Digital Code, the state has created a coordinated, risk-based framework that balances innovation with accountability and digital rights.

Its durability will depend on enforcement and continued capacity development, but the model positions Kazakhstan as a potential regional leader in responsible AI governance. The next publication will examine how artificial intelligence is being integrated across other sectors and ministries, what this development means for investors, and which legal and regulatory risks may arise – together with practical approaches to mitigating those risks.

Footnotes

1. Ministry of Labour and Social Protection of the Population of the Republic of Kazakhstan, ‘90% of Public Services in the Social and Labour Sector Provided in Electronic Format' (Gov.kz, 2025)
https://www.gov.kz/memleket/entities/enbek/press/news/details/831629?lang=ru

2. AfricaNews, ‘AI Adoption Rises in Governments, but Citizen Satisfaction Lags' (AfricaNews, 1 February 2026) https://www.africanews.com/2026/02/12/ai-adoption-rises-in-governments-but-citizen-satisfaction-lags/

3. Fox-Sowell S, ‘AI Can Help Government Speed through Backlogs, New Accenture Report Claims' (StateScoop, 20 November 2025) https://statescoop.com/ai-can-help-government-speed-through-backlogs-new-accenture-report-claims/

4. Benson R, ‘Skill Gaps Barrier to AI Adoption' (Government News, 3 February 2026) https://www.governmentnews.com.au/capability-gaps-barrier-to-ai-adoption/

5. Decree of the President of the Republic of Kazakhstan No 997 of 18 September 2025 ‘On Measures for the Further Improvement of the System of Public Administration' (Republic of Kazakhstan)

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.