AI Data Center Support Should Reflect Market Demand
The question of supporting AI data centers should be considered in the context of both actual market demand for this type of infrastructure and the state of the commercial data center market, says Andrey Guk, CEO of OBIT.

The regulatory classification of data centers into AI data centers and other facilities could enable more targeted support measures. However, according to OBIT CEO Andrey Guk, their effectiveness will depend on how closely they match current market conditions: “Today, the real customers for AI data centers are primarily companies developing their own large language models. There are only a handful of such projects in Russia, so it is premature to speak of an established market and demand for AI data centers.”
Most companies, meanwhile, deploy infrastructure for AI projects either within their own IT environments or at commercial data centers. Such projects require significantly less computing capacity than public AI models handling millions of unique requests per second and typically involve 5-20 high-load racks rather than the construction of dedicated data centers: “If a company does not have its own artificial intelligence model, it does not need an AI data center.”
According to Guk, targeted support for projects focused on developing domestic AI models and infrastructure could be effective in strengthening technological sovereignty. At the same time, the commercial data center segment, which provides capacity for a broad range of businesses and large corporations, should not be overlooked.
Commercial data centers are currently being built on market terms. Operators have no special tax or lending incentives and no preferential electricity rates. The launch of new projects is constrained by limited availability of power capacity and a shortage of suitable sites in the most sought-after regions, including the Moscow area and St. Petersburg. According to CNews Analytics, the volume of new data center capacity commissioned in 2025 fell by threefold compared with the previous year.
“Classifying data centers is certainly an important initiative if it becomes the basis for clear support mechanisms. Also, it is important to consider the market’s actual needs: commercial data centers currently generate most of the demand and provide infrastructure for thousands of Russian companies. Their development is equally important to the country’s digital economy,” Guk said.





![Astra AI [Code] vs. Open Source: Why Deploying an Open Model in a Secure Environment Is Costly and Risky](https://storage.yandexcloud.net/itrussia/uploads/c3b5b6da-1bf7-49cb-817e-0bab1cea8cff.webp)


































