On 10 April 2025, a follow-up workshop titled “From Insight to Action—Applying TAM Outcomes on AI in Kyrgyz Higher Education” was held in Bishkek. Building directly on the Technical Assistance Mission (TAM) seminar “The Role of Artificial Intelligence in Higher Education” (9–10 April 2025), the event convened 20 participants, including representatives from the Ministry of Education and Science, ten leading universities, donor agencies, and private-sector partners.
The workshop focused on two key objectives:
1. Institutional Action Plans
2. Faculty Development Design
The workshop concluded with a jointly endorsed communiqué outlining the current status of AI adoption in Kyrgyz higher education, summarising key recommendations, and assigning follow-up responsibilities with clear deadlines.
This follow-up event marks a critical step in aligning institutional initiatives with the national roadmap on AI, fostering collaboration across sectors to ensure Kyrgyzstan’s higher education system is equipped for the digital age.
As a result of the Follow-up Session, Rector of Adam University – Svetlana Sirmbard had conducted the analysis of the Artificial Intelligence usage in Kyrgyzstan.
The analysis:
Artificial intelligence in Kyrgyzstan is still emerging, yet its trajectory already hints at the country’s digital future. A handful of government pilots, a small cluster of start‑ups, and a series of university experiments reveal genuine momentum, but they also expose the absence of a unifying national vision, modern data infrastructure, and a sufficiently deep talent pool. The most ambitious work is taking place in higher education: Adam University has begun testing a round‑the‑clock “cyber‑mentor” that guides students, analysing academic records with predictive algorithms, and rewriting degree plans to include data‑science and machine‑learning courses. Similar conversations are under way at Osh State University and the American University of Central Asia, yet fewer than one school in three across the country can boast even a basic STEAM laboratory, and most universities admit they urgently need to retrain faculty in AI tools and methods .
The policy environment reflects these contradictions. Kyrgyzstan’s flagship programmes—“Digital Kyrgyzstan” and Taza Koom—refer to AI only in passing, leaving questions of data governance, ethics, and sectoral road‑maps to isolated ministerial initiatives. This strategic vacuum complicates investment decisions and discourages large‑scale experimentation, even as the government signals a willingness to partner with the private sector and maintains a relatively light regulatory touch that makes rapid prototyping possible. Infrastructure lags behind ambition: outside Bishkek, fewer than thirty‑five percent of households enjoy fixed broadband, national cloud services are embryonic, and high‑performance‑computing clusters are non‑existent. Public datasets remain scattered among ministries, rarely machine‑readable or linked, and Kyrgyz‑language corpora—vital for chatbots, voice assistants, and translation engines—still occupy gigabytes rather than terabytes.
Human capital presents another bottleneck. Enrolment in computer‑science programmes is rising, but less than ten percent of graduates specialise in machine learning or data engineering. A 2024 survey at Adam University showed seventy percent of lecturers want formal workshops on AI, while half would welcome long‑term mentoring; at the same time, many of the country’s most promising researchers relocate to Kazakhstan or Russia, further depleting domestic expertise. The nation’s annual scholarly output in AI rarely reaches double‑digit articles in international journals, and doctoral funding remains minimal.
Sectoral adoption is uneven. In public services, simple chatbots answer routine passport or tax questions, but they struggle with Kyrgyz‑language queries and integrate poorly with legacy databases. Two commercial banks employ AI‑driven credit‑scoring models, and the leading mobile operator relies on predictive maintenance for its network, yet such use‑cases remain the exception. In agriculture and environmental management, donor‑funded proof‑of‑concepts use satellite imagery to forecast crop yields and monitor pastures, but these projects seldom progress beyond pilot status.
Several structural challenges impede progress. The lack of a national strategy fragments efforts and blurs institutional accountability. Data scarcity and weak protection frameworks hinder the development of robust local models, while patchy connectivity and the absence of a national GPU cluster stifle computational capacity. Funding for AI research barely exceeds one million US dollars a year, and early‑stage venture capital is scarce. Ethical considerations—algorithmic transparency, bias mitigation, and personal‑data safeguards—are yet to be codified, undermining public trust.
Yet opportunities abound. By embedding AI directly into new e‑government services, Kyrgyzstan could bypass legacy systems and deliver citizen‑centric platforms from the outset. Local‑language natural‑language‑processing would widen access to state information and financial services, while AI‑enabled agri‑tech could boost yields in remote regions and reinforce climate resilience. Personalised learning tools, already piloted at Adam University, offer a route to equalising educational quality and addressing chronic teacher shortages.
Unlocking these benefits will require decisive action. The country needs a five‑year national AI strategy, drafted by an inter‑ministerial task‑force, that sets clear priorities, defines ethical guardrails, and earmarks dedicated funding. A national open‑data portal with machine‑readable APIs, broadband incentives for telecom operators, and a shared GPU cluster for universities and start‑ups would create the technical backbone for innovation. Graduate scholarships in AI, mandatory digital‑literacy modules for teacher‑training colleges, and a diaspora mentorship programme could shore up human capital, while a state‑backed AI innovation fund and regulatory sandboxes in fintech and agri‑tech would stimulate private investment. Proven educational pilots—such as Adam University’s cyber‑mentor and predictive‑analytics dashboard—should be replicated in regional universities, accompanied by the creation of a national repository of Kyrgyz‑language datasets. Finally, a code of ethics covering algorithmic fairness, data privacy, and accessibility must underpin all public‑sector deployments, ensuring services are available in Kyrgyz and Russian and remain inclusive for people with disabilities.
Kyrgyzstan thus stands at a crossroads. The country has taken its first tentative steps toward AI adoption, but turning isolated experiments into nationwide impact will demand strategic clarity, sustained investment, and rigorous attention to ethics and inclusion. With coordinated action on policy, infrastructure, skills, and research, Kyrgyzstan can harness artificial intelligence to modernise education, diversify its economy, and deliver more responsive public services. Failure to move swiftly risks deepening the digital divide and squandering a generational chance to shape an equitable, innovation‑driven future.
