Russian business at a crossroads: Is AI a panacea or an expensive illusion?
The integration of artificial intelligence into business processes has become a global trend, and Russia is no exception. However, as my analysis shows, many companies face a harsh reality: the cost of AI integration often exceeds the expense of maintaining a full-time specialist. Let's examine the pitfalls domestic projects encounter and why regular employees often turn out to be "cheaper" than neural networks.
When honestly calculating the cost of AI implementation, it's not just licenses and tokens that are considered. This includes infrastructure, security, integration with existing systems, staff training, and ongoing support. The main challenge for companies is organizational complexity: how to safely embed AI into the internal network, comply with information security requirements and regulations, without spending years building infrastructure from scratch.
At the same time, the return on such projects can be enormous. In the most radical cases, automation can reduce a department's headcount from dozens of employees to just a few without losing service quality. But the key question is: how to measure this return?
Metrics of Success: Not a "Speed Toy," but a Business Tool
We view AI not as a "toy for response speed," but as a tool that should improve operational and financial performance. The return is measured through two layers: the infrastructure layer (productivity growth and reduced operating costs) and the business layer (how much faster and cheaper a company can launch AI services for internal and external users).
Significantly, we do not observe mass layoffs due to AI implementation. On the contrary, companies are redistributing efforts: spending less on building and maintaining low-level infrastructure and more on creating specific AI scenarios for the business. This changes the profile of tasks for IT teams but does not lead to direct layoffs. By the turn of the year, AI will become for many not a replacement for existing specialists, but a way to enter areas where new staff positions previously had to be created.
Staff Resistance and Skill Degradation: Myth or Reality?
Skepticism from IT teams and the business is a normal reaction to any technology that affects responsibilities and processes. Successful implementation begins not with models, but with a transparent explanation of goals. Employees quickly realize that AI takes over routine operations, smooths out the pace of daily work, and during peak seasons allows them to complete more tasks in the same amount of time.
We do not observe skill degradation where AI is implemented as part of a controlled infrastructure, rather than as a "black box," and is accompanied by training and quality standards. On the contrary, IT specialists gain access to a more modern set of technologies—Kubernetes orchestration, GPU clusters, AI agents—and are forced to grow their competencies to manage these effectively.
Security and Regulation: Playing by Your Own Rules
In our projects, we initially assume that generative models can make mistakes, and we structure solutions so that critical decisions remain with humans. We bet on an in-house platform with a transparent architecture and a manageable perimeter. Given regulatory requirements and the expected tightening of approaches to cross-border data transfer, infrastructure and models must be deployed so that the company can transparently answer where and how its data is stored.
The current situation with the almost complete absence of AI regulation in Russia is an intermediate stage. On one hand, it provides freedom for experimentation; on the other, it creates uncertainty about responsibility, especially regarding generative content and data handling. We advocate for a risk-based approach, where requirements depend on the system's level of impact on people and business, rather than being uniformly applied to all AI services.
My conclusion as an analyst: Russian business is at a unique point. On one hand, there is enormous potential for efficiency growth; on the other, there is a need to independently establish frameworks in the absence of clear rules of the game. The key success factor will not be chasing hype, but a pragmatic approach: a clear understanding of return metrics, built-in security, and a willingness to retrain teams. Those who can find a balance between ambition and reality will gain a tremendous competitive advantage within the next two to three years.