16:00, 21 January 2026

Russia Develops AI Strategist Trained to Think at a Global Level

A “digital adviser” built by researchers in southern Russia is designed to think strategically and help humans tackle global economic questions.

Artificial intelligence and humans “think” differently. Can an algorithm be taught to reason strategically? Russian researchers have set out to answer that question.

In Russia’s Rostov region, scientists are developing an intelligent system intended to help analyze economic conditions and forecast crises. The project is being carried out at Southern Federal University. The system is a so-called “digital adviser” based on artificial intelligence that incorporates elements of human strategic thinking.

More Than Just Statistics

According to the developers, the key distinction between the new system and conventional neural networks is that it does not rely solely on statistics. The AI operates with abstract categories such as political stability, investment climate, levels of trust, and social tension. The “adviser” builds cause-and-effect relationships between these factors. Because such concepts cannot be directly expressed in numerical terms, traditional AI systems often struggle to interpret them.

“We are not teaching the system to recognize patterns in data. We are trying to embed the foundations of human strategic thinking,” said project lead Alexander Tselykh, head of the Department of Information and Analytical Security Systems at Southern Federal University. “When a decision-maker acts, they do not rely on figures from a report alone, but on concepts like stability, investor confidence, geopolitical weight, or social tension. Our system works with exactly these abstractions, building causal links between them the way a human does. The fundamental difference lies in understanding, not pattern recognition.”

Rapid Scenario Modeling

Tselykh stressed that the system is not meant to replace experts. Instead, the AI is conceived as a support tool that can quickly model dozens of scenarios, identify long-term consequences of different decisions, and visualize complex interdependencies. In practice, the new neural network operates at the intersection of multiple domains, from economic policy to crisis management. Training is based on the analysis of texts, including news reports, analytical briefs, and academic publications.

The project is being implemented with grant support from the Russian Science Foundation. At the same time, the university is expanding its educational efforts. In 2025, Southern Federal University launched its first bachelor’s programs under the federal “Artificial Intelligence” initiative, aimed at training specialists to work with emerging technologies.

Science and new technologies