Passions are running high around the development prospects of the voice assistant "Alisa AI." Amid discussions in the Ministry of Digital Development about potentially expanding the functionality of domestic AI assistants on smartphones, tablets, and smart speakers, Yandex has hastened to clarify the situation. Boris Ryutin, head of the security group for "Alisa AI" and the company's autonomous vehicles, categorically denied reports that the assistant could gain unrestricted access to user data, including banking applications.
Official Position of Yandex
According to Ryutin, the capabilities of any assistant on a mobile device are always strictly limited by the operating system's security mechanisms. Access is possible only to those functions and data for which the user has given explicit permission. He emphasized that "Alisa AI" does not receive and will not receive any additional privileges that would place it in a preferential position over foreign counterparts. Moreover, the device owner always has the right to independently choose an assistant—Russian or foreign—in the device settings. This is a fundamental right available to everyone.
The cause for concern was an initiative by the Ministry of Digital Development, under which "Alisa AI" could be included in the list of mandatory pre-installed software, as well as be granted expanded access rights to system functions and applications that collect sensitive data. Market analysts see risks of Yandex monopolizing the digital space in this, but the company itself insists that there will be no "carte blanche."
The Essence of the Legislative Initiative
The draft resolution under discussion, which had already appeared on the portal for regulatory legal acts in early August and was then withdrawn for revision, suggests that the AI assistant could become a "service with access to system functions." This refers to expanded capabilities for working with notifications, authorization, and hardware components. However, as sources familiar with the situation emphasize, even in this version of the document, there was no mention of granting the assistant direct access to banking utilities or the ability to store confidential information on its own infrastructure. All operations were to remain within the scope of permissions granted by the user.
The legislative framework for such changes is already taking shape: in July, the law "On Supporting the Development of Artificial Intelligence in Russia" was adopted, which creates the foundation for implementing a "systemic AI assistant based on sovereign or national foundational models." Experts believe that the main goal of this initiative is to limit the presence of foreign AI models on the Russian market, although there is no direct confirmation of this from the authorities.
My view: The situation around "Alisa" is a vivid example of how regulatory initiatives, even those that are well-intentioned in nature, generate a wave of speculation. It is critically important for the market that any innovations do not violate the fundamental principle of user sovereignty over data. Otherwise, we risk getting not a technological breakthrough, but a new round of distrust toward digital platforms.