Russia weighs easier public procurement of cloud capacity for AI
A government working group is discussing amendments to laws 44-FZ and 223-FZ that would make it simpler for state bodies to buy cloud and computing capacity for AI.

A government working group is discussing changes to Russia's public procurement laws that would make it easier for state bodies and state-owned companies to buy cloud and computing capacity for artificial intelligence. The group, called "Infrastructure and Data Centers for AI", operates under the government's subcommission on AI and is considering amendments to Law 223-FZ, which covers state companies, and Law 44-FZ, which covers public authorities, according to its materials.
The group says current procurement procedures are complex and slow, which limits the use of clouds for critical information infrastructure (KII) in the public sector. Its materials note that exchange-based practice shows such purchases can be much simpler and faster, with guarantees of performance and quality.
What is being proposed
The proposals come from the autonomous non-profit "Coordination Council for Data Centers and Cloud Technologies". It argues that cloud services differ from ordinary goods: they are consumed as needed, capacity can be scaled up or down quickly, and the customer does not buy hardware. The organization suggests several changes:
- Introduce pay-per-use billing for cloud computing and services (IaaS, PaaS, secured clouds) under both 44-FZ and 223-FZ, as well as under standard contracts of the Ministry of Digital Development.
- Simplify supplier selection for secured solutions, allowing a single-supplier purchase if the provider holds the required security certificate, or a simplified selection among providers with a K1-class certificate or a dedicated secured environment for KII facilities.
- Remove the requirement to transfer equipment to the customer's balance sheet, which is irrelevant for cloud services.
Reactions and next steps
The Ministry of Digital Development is not commenting. It said an agreed set of proposals has not yet reached it and any changes will be considered through the established process. The AI Development Center at the government's Analytical Center stressed that talks are at working-group level and that it is too early to speak of adopting amendments.
Dmitry Panyshev of RTK-TsOD, who handles relations with government bodies, said most initiatives are positive but need more detail, and final wording has not been agreed with regulators. He believes that if some proposals are written into law, it would help bring AI solutions into government and business and make tenders easier for customers.
A data-center market source noted that buyers of cloud capacity specify far more than volume: GPUs, platform, support, availability levels (SLA), incident response and data protection requirements. How the procedure is set up directly affects how long it takes to prepare and agree terms before a project launches.
Apps worth a look
Tap the right tiles, add them up, hit the target. Easy to learn, hard to master. Free, for all ages.
Talk to an AI conversation partner; every sentence comes back corrected and explained in your language. 9 languages, A1 to C2.
Open EDIFACT and X12 messages, see every segment explained, convert to JSON/XML, export PDF and Excel. Runs locally, no upload.
Pick who it is for, the occasion and their personality. The search engine reads product descriptions from local shops and ranks what fits. Free, no sign-up.






