AI without illusions: Companies face the risk of rising expectations
Experts warn that the days of “rosy wins” in AI are over, and project economics and algorithmic transparency are coming to the fore.
Hype vs. Economy
Article entitled “Stupidity, greed and deception in direct marketing: the dark side of the “AI” hype” pvsm.ru edition attract attention The problem of overheated marketing surrounding AI solutions. According to the article’s authors, a significant portion of products branded as “AI” (artificial intelligence) are either sophisticated statistical models or automation without full-fledged machine learning.
The author quotes Eugene Kaspersky, founder of Kaspersky Lab. “For many years I was adamantly opposed to the term ‘artificial intelligence’. I explained that it is not intelligence at all. Yes, it is artificial, but it is not intelligence. These are just smart, good algorithms, complex algorithms, but they are algorithms… Strictly speaking, there is no AI, there is machine learning.”
In a very detailed article, the author dissects all the stories surrounding AI and tries to convey a simple idea to the reader. That said, artificial intelligence has contributed to progress and using it is certainly beneficial. However, this does not exempt us from the need to think critically about the role and capabilities of AI and the risks associated with it.
“Today, everything from template-based chatbots to traditional analytics is sold under the name of AI,” said the publication’s authors, highlighting the gap between advertising promises and the actual functionality of solutions.
Meanwhile, global investment in AI will exceed $150 billion in 2025, with companies deploying generative models in HR, marketing, and customer support at scale. However, according to Gartner, up to 30% of AI projects fail to achieve their stated economic impact due to overinflated expectations or a lack of clear ROI business metrics.
This problem is particularly acute for small and medium-sized businesses, where AI implementation is often done without sufficient expertise. As a result, businesses experience increased infrastructure and subscription costs without any measurable increase in revenue.
User and company risks
The main risks are conceptual displacement and lack of transparency. Many solutions use large language models but do not disclose data sources, limitations, or potential errors.
Experts have repeatedly highlighted the problem of “illusory” models. OpenAI CEO Sam Altman said, “AI can generate false information with confidence. Users need to understand the limitations of the technology.” (Source: https://openai.com/Blog).
For businesses, this means reputational and legal risks, from publishing inaccurate data to copyright infringement.
Additionally, the issue of dependence on external platforms remains. When companies build key processes on top of API services, changes in pricing or supplier policies can have a significant impact on the cost of the product.
What this means for the market
The relevance of this topic is related to the transition of the market from the hype stage to the rationalization stage. Investors and customers demand proof of efficiency: reduced costs, accelerated processes, increased conversions.
Key recommendations for AI users include:
– Evaluate economic impact rather than technology.
– Demand transparency of architecture and data sources.
– Test your solution in a pilot project before expanding.
– Consider long-term infrastructure and licensing costs.
The conclusion is clear. Artificial intelligence remains a powerful tool, but its value is determined by measurable results, not what it says out loud. The authors advise that the winners in 2026 will not be those who talk loudly about AI, but those who can do the math.
