Individual use of artificial intelligence may be spreading faster in Hungary than adoption by companies. But economic gains require more than buying new tools: businesses must also redesign workflows, develop employees’ skills and ensure responsible operation, argues Barbora Paulovič Deckerová, president of Microsoft’s Central Europe and Central Asia region, in an opinion article.
Paulovič Deckerová will be the opening speaker at Portfolio’s AI & Digital Transformation 2026 conference on November 26. She has held the regional president role since October 1, 2025.
A gap between individual and company use
Microsoft’s Global AI Diffusion research indicates that approximately one-third of Hungary’s working-age population uses generative AI. Eurostat data, meanwhile, shows that 10.4% of Hungarian businesses with at least ten employees used AI in 2025, compared with an EU average of roughly 20%.
The two measurements are not directly comparable: they cover different groups and types of use. Taken together, however, they suggest that people may be becoming familiar with the technology faster than companies are deliberately integrating it into their operations.
Paulovič Deckerová points to Hungary’s technological achievements – from Neumann János’s work in computing and Gábor Dénes’s theory of holography to Charles Simonyi’s contributions to software – as an encouraging foundation. The next success, she argues, could come from a company able to turn ideas into business results quickly.
Time saved is not an outcome in itself
The opinion article argues that wider chatbot use does not by itself produce economic growth. AI can create value when, for example, an idea can be tested in days rather than months, customer service improves, or a smaller team can compete on a larger scale than before.
Time freed up at work does not automatically translate into financial savings either: its value depends on whether it is used to improve services, reach new markets or resolve customer problems faster. The author draws an analogy with electrification, when the more substantial gains came not simply from replacing steam engines with electric motors but from adapting factory operations to the new technology.
She recommends that companies start with a process important to customers or growth. They should identify bottlenecks and points where better information or additional capacity would help, then redesign the work, measure the results and apply what they learn elsewhere.
Human responsibility and trust
Microsoft uses the term “Frontier Firm” for organizations that deliberately redesign work around collaboration between people and AI agents. That does not mean fully automating every task. AI can analyze information, prepare first drafts or carry out defined steps, but people remain responsible for setting goals and boundaries, evaluating outcomes and making significant decisions.
That distinction is especially important in heavily regulated sectors and work involving customers or sensitive data. According to the author, trust requires appropriate governance, reliable information, clear responsibilities and responsible data handling from the outset.
Training is equally important: employees need to learn to ask better questions, assess answers critically and continually improve their work. The development of human judgment cannot be delegated to AI.
Results in specific processes
The author cites three examples of targeted applications. Bank of Queensland Group shortened a risk-analysis process that previously took weeks to one day with AI; the team checked the results against previous work. At Kuwait Finance House, credit-case assessments fell from three days to about an hour. At Hiscox, identifying and recording the key information in a new insurance claim can take as little as ten minutes, compared with up to an hour previously.
These are results from specific processes, not guarantees that other organizations will achieve the same gains.
Hungarian universities are also examining how AI can support teaching and institutional work. Paulovič Deckerová argues that, alongside access, skills, curiosity and sound judgment will determine how much value the technology can create. Business leaders, she says, must give individual experimentation a secure and reliable place in their organizations and continue improving processes based on what they learn.



