Uzbekistan Wants to Train 5 Million People in AI. Can It Actually Pull It Off?

Uzbekistan Wants to Train 5 Million People in AI. Can It Actually Pull It Off?

Somewhere in the corridors of Tashkent's Ministry of Digital Technologies, a PowerPoint slide almost certainly exists with a very large number on it: five million. That is how many citizens Uzbekistan plans to equip with artificial intelligence skills by 2030, part of a national strategy so sweeping it touches everything from rice paddies to marriage certificates. President Shavkat Mirziyoyev has staked his modernisation agenda on the idea that a landlocked, post-Soviet nation of 37 million can leapfrog its way into the global digital economy. The ambition is undeniable. The question is whether the scaffolding can hold the weight of the dream.

As someone who has watched similar top-down digital transformation pledges unfold across Asia, I find Uzbekistan's gamble both exhilarating and familiar. The rhetoric is polished, the targets are bold, and the gap between announcement and execution remains the only metric that truly matters.

The Blueprint: Digital Uzbekistan 2030

The centrepiece of the push is the Digital Uzbekistan 2030 strategy, approved by presidential decree on 5 October 2020 and updated with increasingly ambitious targets ever since. The plan rests on five pillars: digital infrastructure, e-government, the digital economy, a national technology market, and IT education. By 2030, Uzbekistan aims to digitise 100% of its public services, increase the digital economy's share of GDP 2.5 times, grow software outsourcing tenfold to $500 million, and push its UN E-Government Development Index ranking even higher than its current 63rd place, a leap of 24 positions since 2018.

The AI training component sits at the heart of the education pillar. Platforms like aistudy.uz have been tasked with training one million AI specialists by 2027 alone, feeding into the broader five-million target. The government has earmarked $50 million for a national AI model and 86 sector-specific AI projects spanning agriculture, energy, healthcare, and public administration. In January 2026, new AI legislation updated the regulatory framework to accommodate these ambitions, establishing clearer rules on oversight, liabilities, and responsibilities for AI deployment across industries.

None of this exists in a vacuum. Uzbekistan is watching its neighbours, particularly Kazakhstan, which declared 2026 the "Year of Digitalization and AI" and has automated 92.6% of its public services. The quiet competition between Tashkent and Astana for the title of Central Asia's digital capital is one of the most underreported tech stories in the region.

What Is Actually Working

Credit where it is due: Uzbekistan has moved faster than most observers expected. The Unified Portal of Interactive Public Services, my.gov.uz, now offers more than 760 services online, with a mobile app hosting over 540. In the first half of 2025, the system delivered 16 million individual services. The Unified Identification System has enrolled 12.8 million users across 850 government resources. The MyID biometric identification system is streamlining everything from banking to bureaucratic paperwork, cutting wait times from hours to minutes.

We are expanding digital services for both citizens and businesses. By 2026, we plan to launch a standard business reporting system that will automatically pull data from existing government systems, minimising manual input and increasing efficiency." — Nurshod Nurkulov, Head of Digitalization Division, Ministry of Digital Technologies

The IT sector now employs around 40,000 people, with exports surging from $170 million to $1 billion in recent years. IT Park Uzbekistan, the country's flagship technology incubator, houses over 2,000 companies and provides fiscal incentives designed to keep talent local. Internet speeds have improved sevenfold overall and 4.5 times on mobile, a material improvement even if the starting baseline was low.

For a country that just two decades ago had some of the most restrictive information policies in Central Asia, these are not trivial gains. They represent a genuine, measurable shift in how citizens interact with the state.

The Cracks in the Foundation

Here is where the opinion sharpens: Uzbekistan's digital transformation story reads better in press releases than it does on the ground. A detailed Asian Development Bank analysis found a persistent "design-reality gap" between what the strategy promises and what citizens actually experience. Transactional e-services, the kind where you can complete an entire process online without visiting a government office, remain "in their infancy." Interoperability between agencies is poor. Citizen engagement with digital platforms, beyond basic information retrieval, is still limited.

The infrastructure challenge is real. While Tashkent and a handful of regional capitals enjoy decent connectivity, rural broadband access and electricity reliability remain uneven. Training five million people in AI skills presumes those five million people have reliable access to a device, an internet connection, and the baseline digital literacy to engage with AI tools. That presumption is generous.

Then there is the question no government official wants to answer publicly: where will these newly trained AI workers go? Uzbekistan's domestic tech market, while growing, cannot absorb millions of AI-literate graduates. The risk of brain drain to Russia, Turkey, the Gulf states, or Southeast Asia is significant, and history suggests that training people in globally portable skills without creating enough domestic opportunity to retain them is a recipe for subsidising someone else's tech sector.

## The Comparison That Should Worry Tashkent

Uzbekistan is not the first country to announce a massive AI upskilling programme and declare victory before the hard work begins. Across the region, ASEAN's own five-million-person AI training initiative has discovered that impressive enrolment numbers can mask thin learning outcomes. Singapore, arguably Asia's most digitally mature economy, is handing 100,000 workers free AI tools while simultaneously admitting that access alone does not equal competence.

The pattern repeats: governments announce eye-catching training targets, funnel citizens through short online courses, count completions as "trained," and then struggle to demonstrate that anyone's actual productivity or employability has improved. Research on the AI crutch effect suggests that superficial AI training may even be counterproductive, creating a generation of workers who can operate tools without understanding them, performing well with AI assistance but poorly without it.

Uzbekistan's programme will succeed only if it moves beyond certificate-counting and toward genuine capability building: longer courses, hands-on projects, mentorship from working professionals, and clear pathways from training into employment. That is expensive, slow, and difficult to put on a PowerPoint slide.

What Would Make This Work

The gap between aspiration and execution is not unique to Uzbekistan. It is the defining challenge of AI policy across emerging economies. But Uzbekistan has a few structural advantages that, if leveraged correctly, could tilt the odds in its favour.

First, its population is young. With a median age of around 28, the demographic window for digital upskilling is wide open in a way it is not for ageing societies like South Korea or Japan. Second, the government's willingness to partner with international players, from the ADB to technology firms, provides access to expertise that pure self-reliance would not. Third, the regional competition with Kazakhstan creates a healthy pressure to deliver results rather than just announcements.

What Tashkent needs to avoid is the trap of treating digitalisation as a branding exercise. The world does not need another national AI strategy that looks impressive at Davos but changes nothing in Fergana. The measure of success should not be how many people completed an online module; it should be how many people's lives materially improved because they acquired a skill the economy actually demands.

Metric	Uzbekistan	Kazakhstan
UN E-Government Index Rank	63rd	28th
Public Service Automation	~60%	92.6%
Internet Users	27.2 million	~19 million
IT Exports (Annual)	$1 billion	~$600 million
AI Training Target	5 million by 2030	450,000 by 2026
National AI Hub	IT Park Uzbekistan	Alem.ai / Astana Hub

The AIinASIA View: Uzbekistan's five-million-person AI training target is the most ambitious digital literacy programme in Central Asian history, and that is precisely why it deserves scrutiny rather than applause. The infrastructure is patchy, the domestic job market cannot yet absorb this many AI-skilled workers, and the gap between portal enrolment and genuine competence remains the weakest link in every national upskilling story across Asia. Yet dismissing the effort entirely would be a mistake. Uzbekistan has moved faster on e-government than anyone predicted, its young population is hungry for opportunity, and the competitive pressure from Kazakhstan could push both nations toward results. The gamble is real. So is the upside.

Frequently Asked Questions

Can Uzbekistan realistically train 5 million people in AI by 2030?

The target is achievable in terms of raw enrolment numbers, as online platforms can scale quickly. The harder question is whether that training will be deep enough to translate into employable skills. Short certificate courses count toward the target but may not produce the competence the economy needs. Success depends on course quality, not just quantity.

How does Uzbekistan's AI push compare to Kazakhstan's?

Kazakhstan is further ahead on e-government automation (92.6% vs roughly 60%) and has a higher UN E-Government Index ranking. However, Uzbekistan has a larger population, faster IT export growth, and a more aggressive training target. The two countries are pursuing different strategies, with Kazakhstan focusing on infrastructure and Uzbekistan betting on human capital at scale.

What are the biggest risks to Uzbekistan's digital transformation?

The primary risks are infrastructure gaps in rural areas, a persistent design-reality gap between strategy documents and citizen experience, potential brain drain of newly trained tech workers to higher-paying markets, and the challenge of creating enough domestic AI jobs to absorb millions of trained workers.

Is Uzbekistan's AI training programme different from similar efforts in Southeast Asia?

The scale is comparable to ASEAN's regional AI training initiatives, but Uzbekistan is attempting it as a single nation with a lower baseline of digital infrastructure. The advantage is centralised decision-making and political will; the disadvantage is that the domestic market and ecosystem are less mature than those in Singapore, Vietnam, or Indonesia.

The Bottom Line

Uzbekistan's digital ambitions deserve more attention than they currently receive from the global technology press. Central Asia is not a region most people associate with AI innovation, and that is part of what makes this story compelling. A landlocked country with a GDP per capita under $3,000 is attempting to train an eighth of its population in AI skills within four years. If it works, it becomes a template for every emerging economy watching from the sidelines. If it does not, it joins a long list of national strategies that generated conferences, certificates, and very little else.

The answer will not come from Tashkent's ministries. It will come from whether a farmer in the Fergana Valley, a shopkeeper in Samarkand, or a university student in Nukus can point to a concrete way AI training changed what they do for a living. That is the only metric that matters, and it is the one no government dashboard can fake.

Drop your take in the comments below.