Early July 2026 marked a turning point for the Russian artificial intelligence industry. Within a few days, four landmark events occurred, collectively shaping a completely new architecture for AI regulation and implementation in the country. The state simultaneously acts as both the legislator and the largest user of the technology.
The main event is the State Duma's approval in the first reading of a high-profile bill on AI regulation. The document introduces two categories of fundamental models: "sovereign" and "national." The former are developed exclusively by Russian legal entities, hosted in domestic data centers, and completely exclude the use of foreign components. The latter allow open-source solutions and are created by Russian companies but on domestic infrastructure. Both categories must comply with Russian legislation and consider "traditional spiritual and moral values."
Importantly, the new rules apply not only to domestic YandexGPT and GigaChat but also to foreign models such as ChatGPT, Gemini, and DeepSeek. The government receives the exclusive right to determine state support measures for IT companies, including financial, property, and information benefits. Mandatory labeling of generated audio and video content is separately stipulated.
State Apparatus Shifts to Neural Networks
Alongside legislative efforts, Deputy Prime Minister Dmitry Grigorenko reported on the digitalization of the state apparatus. Forty AI services are already being actively implemented into the work of government agencies. The pilot phase has been successfully completed, and the technologies are being scaled across all structures. According to the official, automation simplifies routine procedural decisions, and programs process vast amounts of data many times faster than any human.
The Ministry of Digital Development, in turn, announced new platforms for managing communication networks using neural networks. Algorithms will be responsible for dynamic distribution of radio frequency spectrum, predicting network load during peak hours, and optimizing the operation of base stations. Initial operator tests have already shown impressive results: data transfer speeds increased by 20-30%, and signal delays decreased by 50%.
Conflict with the Creative Industry
However, the high speed of the law's adoption has exacerbated the conflict with the creative industry. The bill was introduced to the State Duma on June 25 and reviewed on July 7, leaving market participants virtually no time for preparation. On the eve of the vote, leaders of the Russian Book Union, the National Federation of Music Industry, and the Animation Film Association sent letters to authorities requesting urgent revisions to the document.
They are seriously concerned about the wording that permits the free use of texts and images. According to the law, processing protected works for training models is not considered a violation—AI developers only need to obtain a legal copy or find it in open access. Rights holders rightly fear that such a concession will turn the country into a free hub for training neural networks. Creative unions demand transparent conditions for content licensing and official participation in preparing amendments for the second reading.
My expert analysis: Russia is betting on total state control in the AI sector, which is logical in the context of the current import substitution policy. However, the key risk here is not technical but legal. Issues of revenue distribution and responsibility for AI work results remain off the table. If the dialogue with the market is not constructive, we risk getting not flexible regulation but a bureaucratic brake on the entire industry. Time will tell whether the state can find a balance between control and development.