bg
Cybersecurity
19:06, 16 ноября 2026
views
1

Solar Dozor DLP Update Strengthens Controls on AI Interactions

The platform can block prompts that could lead to confidential data leaks.

One of the core tools for protecting corporate information is a data loss prevention (DLP) system. Put simply, these systems monitor user activity and prevent data leaks. Today, organizations use them across virtually every industry, and vendors continually improve their capabilities.

Meeting the Needs of the Moment

Solar Group has announced the release of version 8.4 of its Solar Dozor DLP system. The update focuses primarily on strengthening security controls over employees' interactions with public AI services and corporate messaging platforms.

According to Solar's research for the first half of 2026, nearly 40% of prompts submitted to public AI services contain sensitive information. Of the information being shared, 41% consists of source code and configuration data; 30% involves personal, financial, and other sensitive information; 18% relates to intellectual property; and 11% includes passwords, tokens, and API keys.

Blanket Bans Are Not the Answer

Solar Dozor 8.4 addresses this challenge by analyzing requests and prompts submitted from employees' workstations. The system supports two operating modes: monitoring and blocking. In monitoring mode, the platform simply records all activity and compiles reports for security specialists. In blocking mode, it stops confidential data from being sent when a user violates security policies and displays an alert on the workstation screen. This immediate feedback helps employees learn which information must not be shared with external services.

“Banning the use of AI systems across the board does not solve the problem. Businesses need to understand which ways of working with public AI services are acceptable and which ones put corporate data at risk. Only by gaining a complete picture of how AI is being used can a company prevent security risks rather than respond after data leaks occur,” says Ali Gadzhiev, Solar Dozor product lead at Solar Group.

Deeper Analysis of Messaging Platforms

The system applies a similar approach to corporate messaging platforms. Depending on the security policy settings, it analyzes outgoing messages in both desktop and web versions of applications. This allows companies to refine their information-sharing rules based on detected incidents and supplement employee training with examples drawn from everyday work.

The latest Solar Dozor update reflects a broader shift in Russia's DLP market. Initially, these systems focused primarily on email. Today, the focus is shifting toward monitoring interactions with AI systems and addressing so-called “shadow AI” – when employees turn to AI tools to make their jobs easier without considering the risks this may pose to corporate data.

How the Platform Has Evolved

Solar Dozor has been under continuous development for several years. Version 8.0 was released in 2024, introducing monitoring across multiple communication channels. The new version also introduced DozorDetective, an investigation module that brings together information about security incidents and related events.. It also added a universal 4D player for reviewing captured data and an AI-powered speech-to-text feature.

Subsequent updates expanded the range of features supported across different operating systems. Particular attention, however, has consistently gone to analyzing messages exchanged through messaging platforms. The March 2026 release of version 8.3 introduced file-storage replication mechanisms to protect against the effects of system failures and ransomware attacks. It also strengthened authentication tools and expanded workstation monitoring.

Solar Dozor 8.4 builds on these existing capabilities. The addition of AI interaction monitoring features reflects growing demands for stronger information security. Looking ahead to 2027, Russian DLP systems are likely to gain additional capabilities for monitoring how employees use generative AI models.

We have seen the Russian information security market evolve over the past two to three years, moving from playing catch-up to becoming one of the most competitive markets in the world. No single category of product is a ‘silver bullet.’ Modern security is built as an ecosystem in which DLP, SIEM, next-generation firewalls (NGFW), and DevSecOps tools work together. This is one of the key issues addressed by the research cluster at the Faculty of Information Technology and Big Data Analysis
quote
like
heart
fun
wow
sad
angry
Latest news
Important
Recommended
previous
next