Russia in five numbers
The five numbers below show where money for AI comes from in Russia and what it is expected to give. Each comes from an open source with a date, and the links are at the bottom of the page.
1. $2.7 billion was directed to AI by Russian IT corporations in 2025. These are the sums companies named publicly; Kama Flow collected them.
2. $32 million of venture investment and 33 deals went to AI and machine learning in 9 months of 2026. According to Venture Guide (Moscow Venture Fund), this is first place among sectors by volume and by number of deals.
3. 52:1 is the ratio of corporate to venture money in AI for 2025, by Kama Flow’s estimate. Experts think the method needs checking, so read the ratio as an order of magnitude, not an exact measurement.
4. $1.9 billion is planned for 2026 in the national project Data Economy and Digital Transformation of the State. AI is one part of it; the whole sum does not go to AI.
5. $95–$150 billion a year is what AI could give Russia’s economy by 2030, according to an industry study of 2025. That is up to 5.5% of forecast GDP.
For scale: the whole Russian venture market in 9 months of 2026 was $68 million in 73 deals, with an average deal of $1 million. The data is preliminary, as of 4 October 2026.
The numbers differ in method and period: corporate figures are for 2025, venture figures for 9 months of 2026, budget figures for 2026. You can compare them by order of magnitude, but you cannot add them into one sum.
Corporations
The main source of money for AI in Russia is the own budgets of large groups. According to a Kama Flow study, Russian IT corporations publicly directed $2.7 billion to AI in 2025.
This money goes to in-house development inside the groups: models, computing power, products for clients. It does not show in venture deal statistics, so the market often underestimates how much is really invested in AI.
Corporate money reaches customers through products: platforms, clouds, ready models. A company pays for them by subscription or by request volume, and these payments are also part of its AI spending, even though investment statistics do not include them.
For comparison, Kama Flow gives the venture figure: about $51 million into AI companies in 2025, which is 32% of the whole venture market of $159 million. In 2021 the venture market was $2.5 billion, and AI got $361 million of it.
The conclusion for business is simple. In Russia AI technology is built mostly by large groups, independent vendors are few, and most of them have no large funding rounds behind them.
A detailed analysis of the numbers, a comparison with past years and the caveats on method are on the page on corporations and venture.
Venture
The Russian venture market shrank to $55 million in the first half of 2026, about 48% less than a year earlier, according to the Moscow Venture Fund. There were 54 deals with 33 investors.
A falling market does not mean there is no money for AI. It means there is less of it, and it is concentrated in a few deals.
AI still holds first place. In 9 months of 2026, AI and machine learning got $32 million in 33 deals, while the whole market got $68 million in 73 deals in the same period.
The largest deal of the first half of 2026: ArchiTech AI raised $13 million in a C+ round from UKM Group. One deal visibly moves the whole sector’s statistics, so average market sums say little about a typical round.
Money comes mostly from private funds: 67% of the half-year volume, or $37 million. Business angels gave 16%, corporations 9%, state funds 8%.
What this means in practice: 33 AI deals in 9 months for the whole country. Rounds are few, and every AI company with investor money is visible, so you can check a vendor’s funding through open sources.
All rounds with amounts and sources are in the deal tracker, and the year’s analysis by niche and investor is in the venture report 2026.
The state
The state invests in AI through the national project Data Economy and Digital Transformation of the State. According to the explanatory note to the draft budget, $1.9 billion is planned for it in 2026, more than $2 billion in 2027 and more than $2.2 billion in 2028.
These sums cover the whole national project: data, communications, digital public services and AI. The figures given have no separate line for AI, so you cannot add them to the corporate $2.7 billion.
In venture the state plays a much smaller role: state funds gave 8% of the money in the first half of 2026, according to the Moscow Venture Fund. For a company, state support comes more often through subsidies and grants than through a share in its capital.
What a company should do: before you put the whole project cost into your own budget, check whether it qualifies for subsidies or grants for Russian AI. Program terms change every year, so check them against the documents in force on the current date.
Which measures are available to business and how to get them is covered on the page on state support for AI.
Effect on the economy
According to an industry study of 2025, AI could give Russia’s economy $95–$150 billion a year by 2030, up to 5.5% of forecast GDP. It is based on a survey of 150 technology directors from 16 industries.
In the same study, the companies with the largest effect spend 13–17% of their IT budget on AI. They estimate AI’s contribution at 6.6% of EBITDA, that is, earnings before interest, taxes and depreciation, today, and at 15.8% in their expectations.
The forecast for the country is built from the results of individual companies. The effect goes to those who count it in money on their own processes, not to those who bought more subscriptions.
The gap between companies with an effect and the rest shows in the same numbers. The companies with the largest effect already get 6.6% of EBITDA from AI and expect 15.8%, while the others still have to find the processes where AI pays off.
Scenarios and a breakdown by segment are in the report on the effect of AI on the economy.
What this means for your company
In a market where corporations spend about 52 times more on AI than venture investors, large groups build their own technology and independent vendors are few. When you choose a contractor, you more often work with a small team than with a company backed by a fund.
What a company that wants its share of the effect should do:
1. Calculate where AI will pay back for you: the processes, the losses in money, the order of deployment.
2. Compare your budget with the study’s benchmark: companies with an effect spend 13–17% of their IT budget on AI. If you spend more but have no metric, the money goes without a result.
3. Before you buy a solution or a share in an AI company, check that the product works on your data, not only in a demo.
4. Ask the vendor who funds it. A round in the deal tracker shows whether the company will have money to develop a year from now.
5. Put a metric into the contract: the sum saved per month and the payback period. Without a metric, a year later you cannot tell whether the company joined the group with an effect.
If you look at AI as an investor rather than as a customer, the rule is the same. Venture investment in AI companies in 2025 was about $51 million by Kama Flow’s estimate, deals are few, and the cost of a mistake in one stake is high. Before you invest, check the company’s product, data and revenue as carefully as before buying a solution.
It is easier to start with the AI solutions map: it shows the processes where AI pays off, the losses in money and the order of deployment.