Governments Move AI Into Public Services
Fourteen countries compare notes in Astana as Kazakhstan targets AI skills for one million citizens within five years.
Updated on Nov. 2, 2026 in Artificial Intelligence
Live Poll
Do you trust government use of AI to serve ordinary people fairly?
Representatives from 14 countries shared experiences of using AI in public administration at a conference in Astana, Kazakhstan. The technology described is in active government use rather than a single announced product, with participants from twelve countries across Central Asia, the Caucasus and the Asia-Pacific region presenting their own solutions and implementation challenges.
Why it matters
The exchange shows AI moving from pilot projects into the operational machinery of public services, with governments comparing legal frameworks, digital identity and cyber resilience rather than only testing models. Kazakhstan's push to pair a Digital Code and a dedicated AI law with national supercomputers reflects a broader trend of states building both regulation and compute infrastructure together.
The figures come from national self-reporting presented at a conference, with 92% of Kazakhstan's public services described as fully digital and one million citizens targeted for AI skills training. Independent verification of these numbers and of the program's results is not reported, and the exact method used to measure service digitization is not specified.
The players
Ministry of Artificial Intelligence and Digital Development (Kazakhstan)
Kazakh government ministry that presented the country's AI development experience and oversees its digital governance and AI law.
AI Government Lab (Republic of Korea)
South Korean government initiative where civil servants identify challenges, test AI solutions and develop prototypes for public services.
alem.ai
International AI center opened in Astana, Kazakhstan, serving as a regional hub for AI research and development.
The details
Participants examined how South Korea's AI Government Lab works, in which civil servants identify problems, test AI tools and build prototypes before wider deployment. Kazakhstan described a stack of supporting pieces: a Digital Code, a dedicated AI law, national supercomputers that supply the computing capacity to train and run models, and the alem.ai international AI center in Astana. An interactive session then asked how governments might begin deploying AI agents, which are software systems that carry out multi-step tasks on their own, inside public administration.
Timeline
Conference held in Astana, with the core event dated 2 November 2026.
Kazakhstan's goal of equipping one million citizens with AI skills spans the next five years.
The Tech Race
South Korea's AI Government Lab, where civil servants test and prototype AI tools before rollout, now serves as a reference model for other states setting up similar public-sector AI programs. Kazakhstan's combination of a Digital Code, a dedicated AI law and national supercomputers positions it as a regional contender building both regulatory and compute capacity at once.
Citizens in the countries involved may see AI used more often inside government services, though the source gives no timeline for when public-facing tools will arrive. Kazakhstan's goal of training one million people in AI skills within five years is the only concrete date for personal benefit the source provides.
The takeaway
The trajectory to watch is whether AI agents move from exploratory sessions into deployed public services, a shift the conference only began discussing. Track whether Kazakhstan reaches its target of one million AI-trained citizens within five years, which will show how far the program scales.
Further reading
For broader context on how governments are putting these tools to work, see the Artificial Intelligence section.
Source note: This article includes information reported by UNDP.
Live Poll
Do you trust government use of AI to serve ordinary people fairly?
