South and Central Asia Steps Up AI Readiness as Governments Move From Plans to Practice
Across South and Central Asia, AI plans are moving off the page and into delivery. The latest Government AI Readiness Index, released in December, shows governments investing in compute, skills, and public-sector use cases-with momentum clearly building.
The index, produced annually by Oxford Insights, evaluates how well governments apply AI for public benefit. It covers 195 countries using 69 indicators across 14 dimensions and tracks six pillars: policy capacity, governance, AI infrastructure, public sector adoption, development and diffusion, and resilience.
Regional snapshot: the numbers to watch
- India ranks 27th globally (score: 66.55), leading the region. Recent moves include new AI governance guidelines, a planned national AI Governance Group backed by technical experts, roughly $1.3 billion for AI infrastructure over five years, and hosting the 2026 AI Summit. The country is leaning on strong digital public infrastructure and a deep talent base to advance the INDIAai mission. Learn more.
- Türkiye sits 53rd (58.91). While some earlier strategy items remain pending, the country maintains solid digital and innovation foundations. A dedicated technology visa for AI talent is a promising signal.
- Kazakhstan rises to 58th (56.70), up from 72nd last year-a notable climb. The launch of its first domestic supercomputer expanded national computing capacity and sent a clear message on infrastructure.
- Uzbekistan is pushing forward quickly, targeting five million AI specialists by 2030 and embedding AI across public administration.
Oxford Insights describes the region as a vivid snapshot of global AI progress: active policy work, accelerating infrastructure, and fast-growing talent programs all happening at once.
What this means for public-sector leaders
Government services are the proving ground. The region's long-running focus on e-Government is paying off as ministries pilot AI for service delivery and internal operations. The next leg of the journey is about scale, guardrails, and measurable outcomes.
Success won't happen overnight. Countries that align strategy, workforce skills, infrastructure, and cross-border cooperation-and keep them aligned-will see the most durable results.
Skills are the hinge
Kazakhstan set a 2025 goal to upskill one million citizens in AI basics and advanced capabilities. The new Alem AI hub includes a TUMO center for teens (12-18) with free tracks in animation, programming, 3D graphics, and generative AI. It also hosts Tomorrow School, a free two-year, peer-to-peer AI program emphasizing practice, teamwork, and critical thinking across languages like Golang, JavaScript, Python, and Rust.
Uzbekistan mirrors this ambition with a plan to train five million AI specialists by 2030, while positioning AI as a core tool for public services. The message is clear: talent depth is now a national capability, not a side project.
If you're planning government-wide upskilling, map roles to competencies first, then fund pathways at scale. For structured options by role, see curated tracks here: Courses by Job - Complete AI Training.
Infrastructure and governance are catching up
India is backing policy with money and institutions: fresh guidance, a national governance group supported by technical experts, and multi-year funding for compute. Türkiye's talent visa and Kazakhstan's supercomputer underscore growing focus on capacity and attraction. Uzbekistan is threading AI into core administrative processes.
Across the region, governments are refining procurement, model assessment, and risk management. The aim is consistent guardrails that let agencies move faster with less ambiguity.
The global context: two hubs, many paths
The AI scene is strongly influenced by the U.S. and China, from access to chips to standards. Even so, South and Central Asian governments are charting their own routes-balancing domestic capability with international partnerships.
The report notes that China's 2024 AI research output matched the combined publications of the U.S., U.K., and EU, with 156 institutions each producing over 50 AI papers. Models such as DeepSeek and Alibaba's Qwen showed competitive results at lower development costs, challenging assumptions about hardware needs.
Many countries are mixing strategies: building local capacity, running sandboxes, and launching cross-border research, while also exploring greater technological self-reliance. Expect more experimentation-and a few surprises.
What's ahead
With the G20 in the United States, the G7 in France, and the Global AI Summit in India on the calendar, countries at very different stages will be in the same rooms. We may see a clearer split between "model makers" with the depth to build frontier systems and "model takers" focused on safe, effective deployment.
Five moves to prioritize this quarter
- Pick high-yield use cases: target services with measurable outcomes-faster permits, better benefits accuracy, reduced call volumes. Assign an owner, timeline, and budget.
- Publish an AI governance playbook: risk tiers, procurement checklists, model evaluation criteria, incident response, and audit logs. Keep it simple; update quarterly.
- Secure compute and data: plan for shared GPU capacity and protected data access. Consider a hybrid approach (domestic data center plus cloud) with clear exit options.
- Fund the talent pipeline: scholarships, mid-career reskilling, and university partnerships. Set targets per ministry and report progress publicly.
- Open the door for builders: fast-track visas, sandboxes with clear rules, and standard data-sharing agreements to speed pilots while managing risk.
Sources and further reading
Government AI Readiness Index by Oxford Insights: Report overview
INDIAai mission: indiaai.gov.in
Your membership also unlocks:
