I recently read an interview with one of DeepL’s founders. It offered an interesting and rather different perspective on how AI is developing in Europe. When people talk about artificial intelligence, the conversation often quickly moves to Silicon Valley. The names are familiar: OpenAI, Google, Microsoft, Anthropic or Meta. Europe is frequently described as a continent that regulates technology rather than creates it.
I have always found that narrative too simplistic. Europe may not produce dozens of giant consumer technology platforms, but it does have companies that build highly advanced technology with a very different approach. A good example is DeepL, the Cologne-based AI company that started with machine translation and has grown into one of Europe’s most successful artificial intelligence companies.
In an interview with Bechtle, DeepL founder and CEO Jarek Kutylowski explains why the company has taken a different path from many AI competitors. His message is interesting because it is not about creating the loudest AI story. It is about building technology that people can trust and actually use.
Quality before hype
One of the most striking points from Kutylowski is his focus on quality. DeepL did not become successful by trying to be everywhere at once. Instead, the company concentrated on one very specific problem: making translations better. That sounds simple, but it is exactly where many AI projects struggle. The current AI landscape is full of impressive demonstrations, prototypes and experiments. However, moving from an exciting demo to reliable business use is a completely different challenge.
Kutylowski argues that companies need AI systems that deliver consistent results, especially when they are used in professional environments. A translation error in a casual conversation may be annoying. A wrong translation in a legal document, medical communication or international business process can have much bigger consequences. This focus on reliability is one of the reasons why DeepL has built such a strong reputation.
An opportunity in AI
The success of DeepL also challenges the idea that Europe cannot compete in artificial intelligence. Europe has a different technology culture. Issues such as privacy, security, transparency and control are often considered important from the start. For many organisations, especially in sectors like healthcare, government, finance and manufacturing, these are not secondary concerns. They are fundamental requirements.
In the interview Kutylowski highlights that AI adoption is not only about having the biggest models. It is about solving real problems for users and organisations. That is an important lesson. The future of AI will probably not belong only to companies that create the largest models. It will also belong to companies that understand specific industries, workflows and customer needs.
AI becomes part of everyday work
Another interesting point Kutylowski makes in the interview is the shift from AI experiments towards practical applications. Many organisations are currently exploring AI, but the real challenge is integration. How do you connect AI with existing processes? How do you make sure employees can use it effectively? How do you protect sensitive information?
This is where enterprise AI becomes very different from consumer AI. Businesses need control, security and predictable outcomes. DeepL’s move beyond translation into broader AI-powered communication tools reflects this change. AI is becoming less of a separate technology and more of an invisible layer within everyday business activities.
A more balanced view of artificial intelligence
The interview with Jarek Kutylowski offers a refreshing perspective in a time where AI discussions are often dominated by extremes. Some predict that AI will transform everything overnight. Others focus mainly on the risks. The reality is probably somewhere in between.
AI will change the way organisations work, but success will depend on thoughtful implementation. Companies need technology partners that understand not only algorithms, but also business processes, security requirements and human behaviour. DeepL is an interesting example because it shows another possible path for European technology: focused, specialised and built around trust.
The AI race is not only about who builds the biggest model. It is also about who builds technology that people are willing to use every day. And that is an area where Europe may have more strengths than many people realise.