Kazakhstan Launches AI-Sana to Train 450,000 Students and Teachers in AI Skills

Kazakhstan Just Launched the Most Ambitious National AI Skills Programme in Central Asia

When Kazakhstan declared 2026 its "Year of Digitalization and Artificial Intelligence," it was easy to read it as political branding. The AI-Sana initiative, launched formally in March 2026, suggests something considerably more substantive. The programme targets at least 450,000 students and teachers across schools, universities, and adult retraining tracks — one of the largest coordinated AI skills pushes in Central Asia's history, approved directly by Prime Minister Olzhas Bektenov.

Eight Blocks, Ten Initiatives

The scale of AI-Sana reflects the comprehensiveness of its design. Rather than a single curriculum rollout, the programme structures itself around eight priority blocks and ten key initiatives, each targeting a different segment of Kazakhstan's population.

The components include: regional expansion of the AI School Tomorrow School for specialist training; a new stage of AI University; AI Day Qazaqstan events to build public awareness; TUMO Centres for secondary education; AI Governance^✦ 500 for public sector digital leaders; AI Qyzmet for government employees learning to delegate routine tasks to AI; and AI Corporate for company staff. The official launch took place at Alem.Ai, Kazakhstan's state-backed AI development hub.

Why Kazakhstan Is Moving This Fast

Kazakhstan sits at a peculiar junction in Central Asia's digital landscape. It has the region's most developed AI policy framework, a functioning national AI strategy, and its own large language model (KazLLM) — but it also has a workforce that, outside Almaty and Astana, lacks the practical AI skills needed to leverage^✦ those assets. AI-Sana is an attempt to close that gap at speed.

The urgency has an economic dimension. Kazakhstan's government has set ambitious targets for AI's contribution to GDP growth over the next decade, and those targets require not just infrastructure and models but humans who know how to use them. G42 and FPT's billion-dollar AI cloud deal in Vietnam illustrates how rapidly AI infrastructure investment is moving across Asia — but infrastructure without workforce capability is a wasted asset.

Training 450,000 people in AI skills is not a curriculum project. It is a national transformation programme. Kazakhstan has understood something that many governments are still debating: you cannot buy your way to AI readiness. You have to train your way there.

Comparing Approaches Across Asia

Kazakhstan's top-down, government-designed approach contrasts sharply with how other Asian nations are delivering AI education. The Philippines' DepEd Order 003 mandated AI tool use across all public schools but left significant implementation flexibility to individual teachers and school leaders. Singapore's Committee for AI in Higher Education focuses on governance and standards rather than curriculum delivery. Google and True Corporation's Thailand programme is a private sector-led initiative with government endorsement.

Kazakhstan's AI-Sana is closer to the Japanese and South Korean model: state-designed, state-funded, and delivered through both the education system and direct workplace training.

The Government Workforce Angle

One of AI-Sana's most distinctive elements is its focus on government employees. AI Qyzmet — the track designed for public sector workers — aims to teach civil servants how to delegate routine tasks to AI tools, freeing human time for higher-order decision-making. AI Governance 500 targets 500 digital leaders in government departments specifically.

This is a meaningful distinction from most AI education programmes, which focus on schools and private sector workers. Kazakhstan is explicitly trying to transform how its bureaucracy functions — using AI to improve public services while simultaneously upskilling the people who deliver those services.

The approach reflects President Tokayev's digitalization decree, which frames AI not just as an economic driver but as a tool for government modernisation. The e-gov portal and Kazakhstan's own digital identity systems have already laid the groundwork for AI integration in public services. AI-Sana is the human layer on top of that infrastructure.

Central Asia is watching Kazakhstan closely. If AI-Sana delivers measurable results in workforce capability within 12 months, we will see similar programmes in Uzbekistan, Kyrgyzstan, and Tajikistan.

Challenges Ahead

The ambition of AI-Sana is clear. The execution challenges are equally significant. Kazakhstan's education system outside major cities has significant quality variation, and rolling out AI skills training to rural schools without reliable broadband infrastructure requires solutions that urban-focused programmes rarely address.

Language is another variable. Kazakhstan is officially bilingual (Kazakh and Russian), but AI tools — including KazLLM — are still maturing in Kazakh-language capability. Teaching students and government workers to use AI effectively in their working language, rather than defaulting to English or Russian, is a genuine pedagogical challenge.

NTU Singapore's eight AI training programmes offer a comparison point: structured, credential-bearing programmes that can be independently verified. Kazakhstan's AI-Sana will need similar credentialing frameworks to ensure training translates into verifiable skills — and to attract employer investment in the graduates of these programmes.