Insights
Czech Companies Hesitate With AI: How Do We (Not) Know How to Use AI – And What Can Be Done About It?
September 3, 2025
While the world is moving forward rapidly in the field of artificial intelligence, Europe often maintains a more cautious approach. In many respects, the Czech Republic lags even further behind. We spoke with Ondřej Vaněk, Head of the AI Division at Adastra, about why the adoption of artificial intelligence is slower here, what holds companies back from deploying it, and where AI brings the greatest benefits for industry. In the interview, he shares concrete hands-on experience, shows what really works, and points out what the Czech environment lacks in order to make bigger progress. Available analyses show that Europe is adopting artificial intelligence at a slower pace than the United States. What may be even more surprising is the fact that the Czech Republic is below the EU average.
Interview
How do you explain that?
The European approach to new technologies is generally more cautious than the American or Chinese one. While outside Europe the pressure is on quick results and business success, in the EU the emphasis on protecting fundamental rights, transparency, and oversight plays a major role. A typical example is not only GDPR, but also the currently valid AI Act, which regulates the development and use of artificial intelligence across the Union. Although this regulatory framework introduces stricter rules and may slow down technological adoption, it also creates a stable and responsible environment where the interests of individuals are better protected. Ultimately, it is about finding the right balance between the speed of innovation and social responsibility.
And how does the Czech Republic stand in this regard?
We aspire to be part of the technological forefront, but the reality is more complicated. Compared to countries like Germany or the Netherlands, we are falling behind in investments, infrastructure, and the connections between companies, universities, and startups. We are a nation of pragmatists and often wait until something proves successful elsewhere before we commit to it. What we lack is greater trust, a willingness to take risks, and a more open collaboration between the research and commercial worlds.
Are there any examples in Europe we could take more than just inspiration from?
Yes, the Netherlands can serve as a good model. In their innovation centers, universities, startups, incubators, and large companies work together. Research is directly connected to the needs of industry, and results are quickly translated into practice. It’s not only about technology, but also about a culture of collaboration and a willingness to share experiences. This is the kind of model we would need to develop here as well.
In which areas do you see the greatest potential of artificial intelligence for Czech industry?
We see the biggest benefits of artificial intelligence primarily in the manufacturing sector. This sector forms the backbone of our economy and is closely tied to exports. The ability to produce efficiently determines whether we can withstand international competition. AI helps, for example, with production planning, maintenance management, preventing breakdowns, or improving logistics. Czech companies often have well-established processes and, at the same time, large amounts of operational data. This creates a favorable environment for introducing advanced systems that can learn, respond, and optimize in real time.
We also hold a strong position in the field of cybersecurity. The Czech Republic is home to companies such as Gen Digital (formerly Avast) or Kerio, which have long ranked among the global technology leaders. In this field, AI is commonly used, for example, to detect unusual behavior in networks, analyze security incidents, or automatically assess risks. Development in this area is progressing very dynamically here, not only within companies but also at universities and research centers.
The energy sector is also gaining more and more ground. Until recently, it was a conservative segment with a slow pace of digitalization. However, thanks to stronger investments in recent years, the situation is changing. Energy companies are implementing smart solutions for consumption management, failure prediction, or renewable energy management. Artificial intelligence is used to process operational data in real time and helps respond to changes in both demand and supply. This shift is evident not only among major players but also in smaller technology companies that bring innovative solutions. Overall, it holds true that AI is most effective where data, clear processes, and the need to react quickly and accurately exist. In this respect, Czech industry has a strong starting position.
According to some statistics, less than 10 percent of companies in the Czech Republic use artificial intelligence. What is the main obstacle? Is it finances, concerns, or a lack of time?
Every statistic always depends on the methodology and the way the survey is conducted. We see the figure of ten percent as a very cautious estimate. From our experience, most companies are actively interested in artificial intelligence. Many of them are already working with the technology, they just don’t label it as an official project. In some cases, it’s pilot tests on a limited scale, while elsewhere the solutions are already deployed in full operation and delivering concrete results.
So it cannot be said that ninety percent of companies don’t use AI at all. The reality is rather that a large share of businesses have already taken their first steps. It just hasn’t been talked about that much yet. Moreover, the situation is evolving rapidly. Tools like ChatGPT or Copilot have significantly contributed to bringing artificial intelligence closer to everyday users. Many people have experienced firsthand that AI can write a message, analyze text, or suggest a solution to a problem. And this immediately raises the question of how to transfer that efficiency into everyday work within a team or a company.
Moving AI from the testing phase into full operation, however, is not simple. It’s not enough just to want it. You need the right infrastructure, prepared data, established processes, and competent people who understand both technology and business. This is often where difficulties arise. It’s not a single specific obstacle, but rather a set of factors that influence each other. There is a lack of technical readiness, strategic vision, or courage to commit to a bigger change. And in some cases, also a lack of time, capacity, or internal confidence that AI can truly help.
It’s important to start where it makes the most sense and where the benefits can be demonstrated quickly. Those very first successes often open the door for further steps.
What approaches to AI deployment do you see most often in companies? And what challenges do they face?
We often encounter two scenarios. The first is a bottom-up approach, where AI adoption starts in a specific team or department. However, they usually struggle with budget constraints and have to prove to management that AI makes sense and will deliver results. In this case, the key is to start with something smaller that can quickly demonstrate value, and only then gradually build a larger strategy. Incidentally, this is something we often help companies with.
The second scenario is a top-down approach, where AI is introduced by management. Here, the problem often lies in resistance within individual departments, since it means extra work. What has proven effective in this case are our “AI Days” workshops, which help people get involved and understand that AI can actually make their work easier.
And then there is the role of the Chief Innovation Officer, but they often stand outside the company’s main structure, which makes it difficult to truly enforce anything from their position.
At Adastra, we often act as a bridge that connects both management and specific departments. We understand their needs while also ensuring that the technology is truly adopted and has a clear business impact. Inside a company, this is difficult to achieve because a person will always lean more toward one side. We conduct the dialogue in parallel—on one side with management, and on the other with the teams that will actually work with AI in practice. This way, we are able to achieve truly tangible results in AI implementation.
What tends to be the biggest problem for companies?
It’s not about one specific issue. Sometimes quality data is missing, other times processes aren’t prepared, or the necessary competencies are lacking. AI needs an environment where it can be scaled and managed. If someone is planning production in Excel, it’s hardly sustainable. Moving to a more robust solution isn’t about small steps, but about a leap. And not every company has the courage or capacity for that. When you add in the resistance to new things that exists in some companies, it’s clear why changes happen slowly.
Why do companies stick to Excel so often?
Because it’s accessible, familiar, and easy to adjust. But once it’s used for planning across departments, it becomes unsustainable. Companies often fear that moving to a more advanced tool, such as AI optimization, will be difficult and time-consuming. But the truth is, we have projects in this area where the investment pays off within just 3 to 6 months.
And I’m not talking only about cost savings—the entire planning process becomes significantly faster and more accurate, the company gains greater flexibility, and planning becomes digitalized and, above all, transparent. Suddenly, everyone can see what is being planned, why, and with what data. So AI doesn’t have to mean a revolution overnight, but rather a gradual improvement of processes that delivers results very quickly.
AI Adoption in the Czech Republic
AI adoption in the Czech Republic is still low compared to Europe and the rest of the world.
Europe falls behind the U.S. in AI Adoption
European organizations trail their U.S. counterparts in AI adoption by 45 to 70 percent.
(McKinsey, October 2024)
Czech Republic in Europe
According to Eurostat statistics, 14% of European companies use AI. In the Czech Republic, it is 11%, which is roughly the European average but, for example, significantly behind Germany with 20%.
What is collaboration with a company that wants to start with AI like?
If a company is just beginning with AI, we often start the whole process with an AI Days workshop. During this, we present what AI can actually do in the given industry today, showcase proven examples, and most importantly—involve the company’s key people. Together, we then generate the first set of AI use cases that make sense specifically for their business. AI Days can then be followed by preparing a strategy and roadmap with priorities, where we clearly define what brings the greatest business value, which projects can be deployed quickly, and which will require larger changes.
And what’s important—we take into account from the very beginning how the projects will be deployable in full operation and scalable. Because experiments are fine, but AI only starts to truly earn or save money for the company once it runs fully in day-to-day operations. If a company is already testing AI but struggling with deployment, we analyze the weak points and set out the next steps. The first month is mainly about understanding the company, its goals, IT situation, and internal barriers. We act as a partner that helps connect everything and get things moving.
Do you also invite other companies to these AI Days?
Yes. Sometimes we invite customers who have already worked with us. We enable them to share experiences with each other. When there is no competitive tension, trust and open discussion emerge. For example, we organized a workshop where one client invited six other companies, and together we analyzed specific projects. Many times, it works better when the client speaks rather than us. We then see our role more in providing context, technology, and support.
At the Digital Czechia conference, it was mentioned that some companies already have functioning solutions, but no one comes to ask them how they achieved it. Is that really the case?
Unfortunately, yes. We often see that Czech branches of multinational companies build innovative systems, but then wait for approval or direction from abroad. If the headquarters is in a conservative country like Germany, the whole process can be delayed by years. Yet Czech teams have the skills and the results—they just lack the space to apply them.
Can you help companies align local efforts with headquarters’ decisions?
Yes. Adastra also operates abroad, so we understand both sides. We know what the headquarters wants, what the local team needs, and we look for solutions acceptable to everyone. For example, we help in cases where a Czech company uses a different technology than the parent company. We teach them how to reach agreement, align processes, and use AI without unnecessary conflicts.
What has been your biggest transformation in AI and digitalization so far?
I definitely have to mention our collaboration with Škoda Auto. It’s one of the examples where we truly advanced both data infrastructure and the use of advanced analytics and artificial intelligence. Thanks to these tools, Škoda managed to optimize production and logistics to such an extent that it saves enormous resources. At the same time, it has increased the reliability and scalability of operations.
What’s important—the system runs nonstop. All the know-how of people who had been planning processes manually for years was transferred into a system that now works automatically. And that doesn’t mean replacing those people. It means their experience continues to live on in the system, and they can move on to more demanding tasks. But it’s not only large companies. We also work with smaller manufacturers, such as GZ Media or Bednar FMT. Even there, we succeeded in setting up production so that planning is no longer dependent on individuals and their manual spreadsheets. The company can grow without the fear of being held back by its own system.
What would be your final advice for companies?
Don’t be afraid. AI is here, and it’s here to stay. It’s normal to have respect for something new, but letting fear paralyze you is a waste. Just as trains once replaced horse-drawn carriages, today we are moving into the next phase. Some professions will disappear, but new ones will emerge. History shows that technologies bring more work, not less. AI helps us work smarter, faster, and with greater purpose. Those who embrace it early can gain a lot.
Examples of Successful Projects
AI optimization in manufacturing: Bednar FMT increased capacity by €1.75 million in annual revenue
Czech agricultural machinery manufacturer Bednar FMT replaced manual production planning with a system built on artificial intelligence. In cooperation with Adastra, the company managed to automate forecasts, use capacity more efficiently, and respond more flexibly to changes in demand. As a result, the company shortened planning cycles from days to hours and increased its annual production capacity by revenue worth €1.75 million.
Hyundai uses AI to plan production in 5 minutes and saves CZK 13 million annually
Hyundai Motor Manufacturing Czech implemented a production and paint shop optimization system based on artificial intelligence. The new solution cut planning time from several hours to just 5 minutes and brought significant savings in material consumption thanks to higher color sequencing in production. The annual savings amount to CZK 13 million, with a return on investment achieved in just three months. AI helped deliver a 74% improvement in the primer layer and a 54% improvement in the topcoat application, leading to lower material consumption, fewer defects, and more efficient production.
