Yandex Rolls Out AI Platform to Protect Online Services from Digital Fraud
The platform detects bots, fake requests, artificial traffic manipulation, bulk booking attempts, and other forms of fraudulent activity. It has already been deployed across dozens of services, including Yandex Food, Yandex Afisha, Yandex Travel, and applications powered by Alice AI.
The platform's primary advantage is its ability to integrate with new services in just two to four days instead of requiring months to develop separate anti-fraud solutions. According to the company, deployment in Yandex Food increased the effectiveness of detecting fraudulent requests by more than tenfold. At Yandex Afisha, it also helped reduce large-scale ticket reservations made by bots to create artificial scarcity.
The project is significant not only for Yandex's own ecosystem but also for Russia's broader IT market. Rather than serving as a point solution, it represents a shared infrastructure platform that reflects the shift by major technology companies from isolated anti-fraud tools to unified security architectures. That trend underscores the growing role of AI and big data analytics in cybersecurity, e-commerce, customer support, and online ticketing services.
The number of fraudulent transactions is expected to decline, giving users fairer access to digital services. For the industry, the platform represents a step toward strengthening expertise in applied AI while improving the resilience of digital platforms. In a broader context, the approach aligns with the global trend toward universal anti-fraud systems capable of protecting multiple types of online services.
Potential Beyond the Ecosystem
The platform's main growth opportunity lies in expanding it across additional Yandex services while training its AI models on a larger pool of anonymized data. The ability to identify fraud schemes that migrate between different services is expected to accelerate adoption across more products.
In Russia, solutions of this type are in highest demand across banking and payment systems, e-commerce, advertising, delivery services, ticketing and travel platforms, and online marketplaces. According to the Bank of Russia, 134.2 million fraudulent transactions worth RUB 13.9tn (approximately $177bn) were prevented in 2025. Even so, the number of successful attacks continues to increase.
The platform could eventually evolve into an external product through B2B offerings or cloud services, although the company has not confirmed such plans. Its export potential is tied to CIS countries as well as markets in Asia and the Middle East, provided the technology is adapted to local requirements and complies with data regulations. Importantly, Yandex already has experience bringing internally developed technologies to the commercial market.
Unifying Protection Across Services
In 2023, Yandex integrated AppMetrica and 24metrics to detect fraud in mobile advertising, including click spam and bot-generated activity. A year later, MTS introduced an anti-fraud platform for banks capable of analyzing subscriber behavior and launched an AI-powered service that warns users about fraudulent activity during phone calls.
In 2025, Yandex strengthened anti-fraud protections in Yandex Direct by blocking part of the invalid traffic before advertisements were displayed. As a result, fraudulent conversions were reduced by half, while fake leads declined by a factor of 3.5. At the same time, total losses from advertising fraud exceeded RUB 10bn (approximately $127m), and the share of sophisticated fraudulent traffic increased to 13%.
Against that backdrop, Yandex's latest initiative stands out because of its scale. Earlier anti-fraud solutions were built for individual business segments, whereas Universal Anti-Fraud (Universalny Antifrod, Universal Anti-Fraud) brings together protection for multiple services within a single ecosystem.
Expanding Fraud Detection Capabilities
Universal Anti-Fraud should be viewed as a core infrastructure project that standardizes protection across Yandex services while significantly shortening deployment times. Its primary objective is to establish a unified platform that enables anti-fraud capabilities to be launched rapidly for specific threat scenarios.
Over the next several years, most Yandex services are expected to join the platform, while the range of detectable fraud schemes will expand and data sharing across products will become more extensive. An external version of the platform could also emerge in the future.
Meanwhile, the competition with cybercriminals will continue to intensify. Cybercriminals are continuously adapting, increasingly leveraging automation and AI tools.. The platform's long-term effectiveness will depend on how quickly its models can be updated, the quality of the underlying data, and its ability to minimize false positives. More broadly, anti-fraud technology in Russia is evolving into a distinct domain within enterprise IT infrastructure.
