AI News: The Day's Highlights

June 22, 2026 · 4 min read

Perplexity

The most notable recent news in this topic is not about another chatbot, but about the rules of the game: the Ministry of Digital Development of the Russian Federation has published a draft law on state regulation of AI, which directly permits the use of AI in state information systems and critical infrastructure facilities only for models from a trusted registry. The date of the event is 2026; the draft also stipulates that citizens must be informed about the use of AI and be given the opportunity to receive services without it.[4]

This is an important turning point for public services: the focus shifts from "implementing AI faster" to control, transparency, and the right to an alternative service scenario. For users, this means more predictable digital services, and for the state, stricter requirements for the quality and security of automated solutions.[4]

Sources:

Ministry of Digital Development of Russia: draft law "On the Fundamentals of State Regulation of the Use of Artificial Intelligence Technologies in the Russian Federation..."

Gemini

A national AI assistant is actively being developed in Russia, which will be integrated into the "Gosuslugi" (Public Services) portal. According to the head of the Ministry of Digital Development, Maksut Shadayev, a prototype is already under development. The project's goal is to create a single point of access for citizens to resolve everyday and governmental issues, leveraging the potential of "Gosuslugi" as the largest platform. This will make interaction with the government even simpler and more convenient.

In the Moscow region, AI is already successfully used to verify documents attached to online applications on the public services portal. The system can recognize over 700 types of documents, allowing it to identify and correct over 90% of errors before the application is submitted to the agency. This significantly reduces the number of rejections and speeds up the processing of applications, saving time for both citizens and agency employees.

The United States is experiencing rapid growth in the use of AI in government agencies, with concerns noted about the lack of transparency in these processes. The list of AI application scenarios has expanded by 70% since the departure of the Biden administration, including the transfer of sensitive government functions to artificial intelligence. This raises questions about the extent of decision-making processes being transferred from humans to machines in areas such as individual freedom, public health, and nuclear reactor safety.

Sources:

Interfax
CNews
3DNews
The Guardian

ChatGPT

In May 2026, artificial intelligence checked over 3.4 million documents on the regional public services portal in the Moscow region. The system analyzes attached files before sending an application to the department, recognizing over 700 types of documents, including birth certificates, powers of attorney, and bank statements. As a result, over 90% of errors are eliminated before submitting an application, and registration time is reduced by approximately one and a half minutes. (ferra.ru)

Grok

On June 2, 2026, U.S. President Donald Trump signed an executive order "On Promoting Innovation and Security in Advanced Artificial Intelligence." The document instructs federal agencies to modernize government information systems with AI tools within a short timeframe (30–60 days), strengthen the cybersecurity of critical infrastructure, and ensure government agencies' access to advanced AI models to combat threats.

The initiative aims to accelerate the adoption of AI in the daily operations of the government apparatus, from automating services for citizens to strengthening national security. This significantly changes the approach: instead of experiments at the level of individual agencies, it involves systemic integration of technologies at the federal level.

Claude

AI Audit Reveals Half a Trillion Rubles in Unjustified Spending in State Budgets

The Ministry of Finance and the Ministry of Digital Development are implementing an AI-based automated system called "Budget Controller" across all regions of the Russian Federation. This is not merely a technical innovation; initial results reveal large-scale embezzlement of budget funds that had gone unnoticed by human auditors for years. According to the Accounts Chamber, the total volume of violations in the use of budget funds in 2025 amounted to 1 trillion 527.6 billion rubles.

In just three months of pilot testing, the system has already provided impressive examples: in the Nizhny Novgorod region, 142 contracts were identified with signs of inflated prices totaling 87 million rubles; in the Sverdlovsk region, duplicate procurements were discovered with a potential saving of 12 million rubles. According to the developers' forecasts, the full-scale implementation of AI auditing could bring budgets at all levels up to 50 billion rubles in annual savings.

It is important to note: algorithms will not replace humans; the final decision remains with the control and audit bodies. The system is planned to be scaled to all regions by October 2026.

Event Date: May 2026

Sources:

Blog