Kazakhstan - AI Development Strategy (No. 592/2024)

Overview

The Concept for the Development of Artificial Intelligence for 2024–2029 (approved by Government Resolution No. 592 dated 24 July 2024) is Kazakhstan’s primary strategic framework for national AI development. It sets out objectives, KPIs and a multi‑year action plan to increase the number of AI products, create national datasets and build infrastructure such as national data centers and a planned supercomputer. The Concept positions the Ministry of Digital Development, Innovations and Aerospace Industry as the lead implementing agency and references the newly established Committee for Artificial Intelligence and Innovation Development as a key operational body. The Concept and its Plan of Actions are published in the official legal information system (Adilet) and summarized on the Prime Minister’s information portal; official texts and the Plan are available at the government legal repository (Adilet, P2400000592) and described in the Prime Minister’s press release (primeminister.kz).

Definitions

The Concept uses a set of operational definitions consistent with national practice for strategic documents. Key terms include "artificial intelligence" (methods and systems that perform tasks typically requiring human intelligence), "national platform" (the state‑backed technical and governance environment enabling development and deployment of AI products), "datasets" (curated, annotated collections of data made available for model training and testing), "supercomputer" (high performance computing resource envisioned to support large‑scale AI workloads), and "stakeholders" (state bodies, research organizations, commercial developers, and regional administrations). The document also defines the roles of implementing institutions and references existing data repositories (Smart Data Ukimet) and the National Information Technologies JSC (AO "NIT").

Governance and Institutional Framework

The Concept establishes a layered governance structure. The Government approved the Concept by resolution and charged the Ministry of Digital Development, Innovations and Aerospace Industry (MDDIAI) with coordination and control — the MDDIAI must submit annual implementation reports to Government by February 1 each year. The Concept references the creation of the Committee for Artificial Intelligence and Innovation Development (established by Government Resolution No. 342 of 27 April 2024) which is intended to operate as a specialized coordinating body to implement state policy, prepare regulatory proposals, and liaise with sectoral ministries and regional akimats. The Prime Minister’s office and central communications units provide political oversight and public communications; the document also envisages close cooperation with state corporations and the National Information Technologies JSC for infrastructure delivery. Governance measures include mandates for interagency coordination, defined responsible bodies for each action in the Plan, and institutional KPIs to be tracked by MDDIAI and reported publicly.

Key Focus Areas

The Concept organizes initiatives across five principal directions. (1) Data — establishing rules and technical standards for quality, unified formats, cataloguing and sharing of government datasets; targets are set for the cumulative number of curated datasets by year. (2) Infrastructure — phased construction of data centers, expansion of fiber and cloud availability, and deployment of national high‑performance computing capacity (supercomputer planned in phases). (3) Human capital — curriculum development, retraining programs, scholarships and large‑scale upskilling (the Concept foresees training programs to increase AI specialist numbers). (4) R&D and innovation — incentives for private‑public research partnerships, funding mechanisms, national AI platform to host tools and solutions, and incubation/commercialization supports. (5) Regulation and governance — recommendations on legal frameworks, conformity, certification, cybersecurity and ethics; the Concept advocates a risk‑based regulatory approach and anticipates subsequent specific technical regulations and standards for safety, testing, transparency and data protection. Sectoral priority areas named in the Concept include public administration, healthcare, energy, transport, mining, agriculture, education and financial services.

Implementation Framework

The Concept is accompanied by a detailed Plan of Actions (attached to the resolution) that assigns specific measures, deadlines and responsible institutions. Each measure in the Plan lists a form of completion (e.g., prepared datasets, launched infrastructure, adopted standards), a timeframe (annual and milestone targets for 2024–2029) and lead implementers (MDDIAI, relevant line ministries, akimats, AО "NIT", and other state bodies). The Plan includes quantitative indicators (datasets, number of AI products on the national platform, investment targets) and phases for infrastructure deployment (data centers, supercomputer phases). Implementation emphasizes public‑private collaboration, leveraging state data assets (Smart Data Ukimet) while proposing model contractual forms and access arrangements for business and research. The MDDIAI is required to coordinate monitoring and produce the annual progress report to Government.

Monitoring and Evaluation

Monitoring is structured around the Plan of Actions KPIs and annual reporting obligations. The Concept requires the MDDIAI to submit an implementation report to the Government annually (due by February 1). The Plan specifies milestone indicators for each year to 2029 (datasets collected, AI products on the National AI Platform, investment and export targets for IT products). Monitoring mechanisms include interagency progress meetings, public reporting of selected indicators, and technical evaluation of infrastructure delivery. The Concept also foresees the development of evaluation methodologies for AI product performance, conformity assessment processes and standards for safety and quality, to be detailed in subsequent normative acts.

Penalties, Liability, and Appeals

The Concept itself is primarily strategic and does not comprehensively list prescriptive penalties; instead it calls for development of regulatory measures to manage risks, ensure accountability and protect rights. Liability and enforcement for misuse or harms caused by AI systems are to be addressed in subsequent legislation and sectoral regulation, drawing on existing administrative and civil liability frameworks. The Concept recommends establishment of conformity and certification procedures and envisages that non‑compliance with future registration/certification or data protection rules will carry administrative sanctions or orders under relevant codes. The document also outlines institutional channels for appeals and redress tied to administrative practice and recommends establishing clear procedures for affected persons to seek remedies once operational regulations are enacted.

Relationship to Other Instruments

The Concept is explicitly linked to and intended to complement existing national strategies, notably the Digital Transformation Concept (2023–2029), sectoral development strategies and the National Plan 2029. The Plan of Actions references and aligns with the Plan for the development of the National Information Technologies JSC and other state programs. It also cross‑references the Government’s decision establishing the Committee for AI (Resolution No. 342) and indicates that many implementation activities will be coordinated with the National Data initiatives (Smart Data Ukimet) and the State investment roadmap. The Concept is therefore a policy layer that orients sectoral reforms and signals the need for follow‑on regulatory acts and technical standards.

International Alignment

The Concept emphasizes international cooperation, benchmarking and alignment with global best practices. It calls for participation in international research cooperation, adoption of recognized technical standards, and harmonization with internationally accepted principles on AI safety, transparency and human rights protection. The document highlights the importance of cross‑border data flows for research and trade while flagging the need to balance openness with national security, data protection and sovereignty. It encourages Kazakhstan to engage in multilateral fora and to align regulatory approaches with peer jurisdictions to facilitate exports of AI products and foreign investment.

Implementation Timeline

Sources and References