Russia’s largest companies and business associations say a bill on state regulation of artificial intelligence would create serious risks for businesses and the public if passed in its current form.
The assessment comes from reviews of the draft published on the regulation.gov.ru portal, the Russian business outlet RBC reported. More than 150 experts took part in the discussion, including representatives of Rosneft, Rosseti, the Association of Computer and Information Technology Enterprises (APKIT), the Association of Digital Platforms, the Association of European Businesses (AEB), the Chamber of Commerce and Industry, MegaFon, the combined Wildberries and Russ company, and others.
If the bill passes as written, participants warned, companies’ costs for implementing AI will rise, products will reach the market more slowly, development will shift to other jurisdictions, and Russians’ access to advanced technologies — such as diagnostic and treatment tools — will be restricted.
The bill’s requirements for “sovereign” and “national” AI models drew the most criticism. Both categories cover models that go through all stages of development and training in Russia, on Russian datasets, and under the direction of Russian citizens and companies.
Rosneft said in its review that the requirement is technically unworkable, noting that relevant Russian-language data in open sources is insufficient and that the necessary computing infrastructure has not yet been built. The Chamber of Commerce and Industry, the AEB, and the Association of Digital Platforms said in their reviews that no models currently operating in Russia meet all the stated criteria for “sovereignty,” since all Russian-developed models rely on foreign components and open datasets. APKIT flagged the lack of clarity distinguishing “sovereign” from “national” models, noting that the core requirements for both are identical in the bill, with no specifics provided for either.
Experts also raised objections to other provisions of the bill, RBC reported, including a requirement to use only “trusted” models that must meet government-set safety and quality standards and process data exclusively within Russia.
The AEB also criticized the bill’s principle of “accounting for and respecting traditional Russian spiritual and moral values.” Concepts such as “high moral ideals,” “the primacy of the spiritual over the material,” and “the strong family” are not legal categories and cannot serve as the basis for legally significant decisions about whether a model may enter the market, the association’s representatives said.
Materials on Regulation.gov.ru indicate that the Digital Development Ministry partially incorporated the market participants’ proposals when revising the bill.
The draft law on state regulation of artificial intelligence in Russia was published by the Digital Development Ministry on the regulatory drafts portal in mid-March. Public discussion of the document ran from March 18 to April 15. The bill has not yet been submitted to the lower house of Russia’s parliament.
At Meduza, we are committed to transparency about our use of artificial intelligence in the newsroom. The story you’re reading was written by one of our living, breathing journalists and translated from Russian using an AI model configured to follow our strict editorial standards. This translation process is the result of extensive testing and refinements to ensure our English-language coverage is timely and accurate. A Meduza editor reviews every draft before publication.
If you find any errors in this translation, please contact us at [email protected].
To read Meduza’s exclusive content in English, please subscribe to our newsletter.
