Artificial Intelligence as a Data Guardian: Researchers from VŠE Are Contributing to a Next-Generation Security System
Security systems can detect suspicious handling of sensitive data, but they do not always correctly determine whether it represents a real threat. The project “Next Gen User Interface Using Local AI” will develop a solution that enables risks to be assessed more accurately and appropriate responses to be selected. The Faculty of Economics at the Prague University of Economics and Business will also contribute to the project by providing expertise in behavioural economics.
The new software will use a combination of cloud-based and local artificial intelligence. As a result, it should be able to better take into account the context of a specific situation, adapt communication and offer recommendations on the next steps. The project aims to reduce false alarms, lower the time burden on users and strengthen the protection of sensitive data in a hybrid digital environment.
The effectiveness of protection also depends on an understanding of human behaviour and the conditions under which people use new technologies. This is precisely the area on which experts from the Faculty of Economics at the Prague University of Economics and Business will focus. The research team will analyse user behaviour and decision-making when dealing with security situations, design the research framework and test user interaction with the new interface in the Laboratory of Experimental Economics.
At the university level, the project will be implemented through cooperation between the Faculty of Economics and the Faculty of Management in Jindřichův Hradec, which will bring an important perspective from managerial practice, organisational impacts and business processes. The project combines the research background of the Prague University of Economics and Business with the experience of partners from technological practice. In addition to the university, Safetica a.s. and FT Park z.ú. will also take part in the project.
Interview with Vojtěch Kotrba, the principal investigator of the project on behalf of VŠE.
What was it about this topic that interested you so much that you decided to lead the project on behalf of VŠE?
What interested me most was the connection between cutting-edge technical practice and behavioural economics. Today’s security software often fails not because it has poor code, but because it ignores the human factor. It overloads and frustrates users, who then bypass security rules. The opportunity to use our Laboratory of Experimental Economics (LEE) to test how real people respond to an adaptive AI interface in real time is a unique research opportunity.
Which part of the research do you personally consider the most challenging, and which part are you looking forward to the most?
The most challenging part will undoubtedly be the design of the experiments themselves. In laboratory conditions, we need to simulate situations that faithfully reflect the stress and routine experienced by employees in practice, while at the same time isolating the influence of individual elements of the AI interface. And that is exactly what I am looking forward to the most: seeing hard data from LEE on how different variants of interaction change user behaviour and how significantly we can reduce their cognitive load.
Do you see any further directions in which this research could develop in the future?
Certainly. The integration of local and cloud-based AI into everyday work tools is only just beginning. Research into user behaviour when interacting with autonomous security systems opens the door to a broader study of human trust in AI and the optimisation of digital environments in other areas where both security and user comfort are critical.
What will the project bring to the Faculty of Economics, its researchers and potentially also its students?
For the faculty, this is an excellent opportunity to show that behavioural and experimental economics has a direct overlap with current cutting-edge technology projects. For our researchers, the project will provide unique data from testing in LEE and the opportunity to publish in prestigious applied journals. Students will be involved both in the preparation and implementation of the experiments themselves and will gain access to real-world challenges from practice in cooperation with leaders in the field of cybersecurity.
*The project Next Gen User Interface Using Local AI (NextGenAI), registration number CZ.01.01.01/01/25_067/0009767, is being developed under the OP TAK Application programme – Call IV – DEEP TECH.