Infosecurity Launches AI Security Services Suite
Russian businesses now have access to dedicated tools for managing AI security risks. Infosecurity Softline Solutions (Infosecurity Softline Resheniya), part of the Softline Group, has introduced a comprehensive service portfolio covering AI-security maturity assessments, AI-system risk analysis, LLM application red teaming (a proactive security-testing practice for large language models), LLMSecOps implementation (the integration of security controls across the entire lifecycle of large language models), development of corporate AI security governance frameworks, and preparation for ISO/IEC 42001 certification.
The launch of this service suite reflects a broader industry trend. It represents the emergence of a specialized B2B practice aimed at organizations deploying neural-network technologies in software development, analytics, customer services, and internal business processes. Notably, AI security is increasingly being treated as a distinct branch of cybersecurity rather than merely an extension of traditional information security.
The expected outcome is more secure use of AI in banking, public services, healthcare, and retail. That could reduce the risks of data breaches and errors in automated decision-making systems. At the national level, the issue is also tied to technological sovereignty as the AI market continues to expand. According to CNews, revenue generated by Russia's top 35 AI solution providers reached 43 billion rubles (approximately $550 million) in 2024, representing growth of more than 80%.
Orientation Toward International Standards
The Infosecurity Softline Solutions initiative has strong prospects within Russia as the country's AI Security market continues to take shape. Key areas include AI-system auditing, LLM application protection, vulnerability testing, model-risk assessment, and data governance.
By the fall of 2025, more than 50 Russian companies were operating in areas related to AI security. However, only 11 specialized exclusively in this field, highlighting the market's early stage of development. These figures were reported by SecPost.
Demand for such services is expected to increase as AI becomes embedded in critical business processes and regulatory requirements continue to evolve. Export opportunities are likely to be concentrated in friendly markets, with the strongest demand centered on methodologies, audit services, and LLM-protection solutions aligned with international standards, including ISO/IEC 42001.
From Discussion to Operational Practice
Russia has spent several years building both the institutional and practical foundations of AI security. In 2024, a specialized consortium was established to protect data and promote trusted AI technologies. A year later, market participants and professional communities formalized approaches to securing LLMs. Positive Technologies published policies for the secure development and operation of AI models, while OWASP released guidance identifying major threats, including prompt injection attacks. Together, these developments signal a shift from discussion to implementation, where organizations require not only technical testing but also formal AI governance frameworks.
In 2026, AI red teaming emerged as a standalone segment of the cybersecurity market. Companies including Group-IB began offering services focused on testing GenAI systems, LLMs, and related infrastructure. Risks such as prompt injection, data leakage, data poisoning, and supply-chain attacks are increasingly being treated as a dedicated category of security concerns.
At the same time, regulatory activity is accelerating. Proposed legislative changes would allow AI models to access government data only with approval from FSTEC (Federal Service for Technical and Export Control) and the FSB (Federal Security Service). The proposals would also formally assign responsibility for AI-system security to system operators.
The Focus Shifts to Risk Management
The introduction of Infosecurity's service suite underscores the evolution of the Russian market from rapid AI adoption toward secure and controlled AI operations. Under this model, the focus for businesses shifts from efficiency gains alone to the management of risks associated with data leakage, vulnerabilities in LLM applications, model failures, shadow AI usage, and compliance with emerging standards.
Over the next several years, demand for AI Security services is expected to grow across financial services, government, industry, telecommunications, healthcare, and software development. As a result, a new segment of the cybersecurity market is likely to emerge around the protection of AI models, data, and the processes governing their use.
