Mongolia wants to become the place where the world stores its data. During the UNCCD COP17 conference in Ulaanbaatar, running from August 17 to 28, the government rolled out a package of incentives designed to lure data center investors: tax relief, cheap energy, abundant land, data security guarantees, and naturally cold weather that slashes cooling costs.
The target is ambitious. Mongolia is aiming to attract at least $2 billion in investment for AI and green data center initiatives, a figure that would represent a meaningful chunk of new capital for a country whose economy has historically revolved around mining and livestock.
The pitch: nature’s air conditioning and sovereign backing
The centerpiece of the strategy is the Hunnu City smart city project, a 31,000-hectare development located roughly 50 km from Ulaanbaatar. The site is planned to run on geothermal and other renewable energy sources, positioning it as a “green” data center zone.
Backing the initiative is the Chinggis Khaan Sovereign Wealth Fund, which currently manages reserves of $1.4 billion. The fund is actively seeking data center developers to participate in the Hunnu City project and related infrastructure.
Mongolia is also developing a national data center strategy covering 2025 to 2030. The plan includes streamlined land allocation procedures, enhanced cybersecurity standards, and the creation of international-standard “free data zones” designed to integrate data flows into the country’s broader economic framework.
The reality check
For all the ambition, the current state of Mongolia’s data center industry is modest. The country has only a handful of Tier II facilities, primarily clustered in Ulaanbaatar and operated by domestic players like Mogul, Gemnet, and Ssystems. There are no foreign hyperscale commitments announced as of the COP17 conference.
Tier II facilities offer basic redundancy but fall short of the Tier III and Tier IV standards that major cloud providers and enterprises typically require.
Mongolia is landlocked between Russia and China. Latency-sensitive workloads destined for users in Seoul, Tokyo, or Singapore would face physical distance constraints. But for AI training, cold storage, and batch processing tasks where milliseconds matter less, the economics could be compelling.
Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our Editorial Policy.

1 hour ago
13









English (US) ·