Artificial Intelligence in Higher Education: From Specific Practices to Systemic Changes

This paper from the University of Ljubljana offers insight into the thoughtful integration of generative artificial intelligence (genAI) into teaching. Asst. Prof. Boštjan Bajec, Ph.D. (Faculty of Arts, University of Ljubljana) has introduced an activity titled “Training in the Recruitment Process” in his Psychology course, which provides students with the most realistic experience possible of evaluating candidates through the various stages of the selection process. Students thus gain practical experience working with the key resources and data that HR professionals encounter: resumes (CVs), HR questionnaires, test results, and candidates’ responses during interviews. AI was used to support student assignments: after submitting each step in the Moodle classroom, the student receives “feedback” generated by MS 365 Copilot from fictional candidates with predefined characteristics (e.g., personality fit, motivation, skills). The process is designed in stages: based on a job analysis, the student prepares 1) a job posting, 2) a recruitment questionnaire, 3) a set of traits to be assessed, and 4) a selection interview; they then receive 1) the CVs of five candidates, 2) the questionnaire responses of four candidates, 3) the test results of three candidates, and 4) interview-like responses from two candidates. After each step, they must eliminate the least suitable candidate, providing a justification, and in conclusion, describe the selected candidate and propose guidelines for working with them. The rules for using the AI are clear: students may use it to prepare individual steps, but they must exercise professional integrity in assessing the accuracy of the AI-generated texts and must also acknowledge the use of the AI in the final report. The exercise raised several questions and prompted reflection, and the generated responses were mostly of high quality, although sometimes somewhat similar in form; in the future, the instructor will describe the candidates in greater detail to ensure more diverse responses. From an ethical and privacy standpoint, this approach is safe, as the candidates are fictional; to protect students, the materials are processed on university servers without attributing authorship of student work in MS 365 Copilot.

Research shows that higher education instructors’ skepticism toward genAI often stems from well-considered concerns about the quality of teaching and learning, not merely from resistance to technological change. Concerns related to the accuracy of information, ethical issues, and the impact on student learning and academic integrity highlight areas that require a systemic approach. The authors emphasize that it is precisely an understanding of these concerns that enables the development of more thoughtful and responsible approaches to the implementation of AI in higher education. Rather than the rapid adoption of individual tools, a gradual approach that includes dialogue with faculty and clearer institutional frameworks for the use of AI appears to be more appropriate.

International experience shows that the effective use of genAI in higher education is most successfully developed in environments where teachers actively share their experiences and practices. The open exchange of case studies facilitates a quicker understanding of both the advantages and limitations of individual tools and encourages reflection on one’s own teaching. It is important that, when introducing new approaches, teachers are not left to experiment on their own, but rather have access to the experiences of their colleagues and supportive structures. Such a community-based approach reduces uncertainty and contributes to a more thoughtful and high-quality use of digital tools in teaching.

Reflections on the use of AI in teaching are also directly linked to the question of what knowledge and competencies students need for the future. Discussions about AI increasingly point out that key competencies for the future go beyond mere technical knowledge. Employers are increasingly emphasizing the importance of contextual understanding, critical judgment, and the meaningful and responsible use of tools, which also includes an awareness of the limitations of AI. At the same time, data show that AI does not currently reduce employment opportunities, but rather is changing the structure of the skills required. Rather than replacing people, the focus is on complementing their abilities, with the continuous development of digital and transferable skills becoming a key factor in long-term employability.

Discussions on academic integrity in the age of genAI increasingly emphasize the importance of instructional design. Rather than focusing on monitoring technology use, an effective approach is to design assignments that require students to demonstrate understanding, reflection, and active participation. Such learning situations reduce the likelihood of inappropriate technology use and encourage greater student engagement in the learning process. Academic integrity is thus increasingly understood as the result of high-quality teaching and thoughtful instructional design.

The role of AI is not limited solely to the learning process; it is also increasingly transforming research work at universities. The development of AI is bringing increasingly advanced and specialized tools that enable a deeper understanding of complex problems and provide significant support for research. Their adaptability to specific research fields also opens up new possibilities for the meaningful integration of research and education. This is confirmed by current research at the Faculty of Mechanical Engineering at the University of Ljubljana, where AI significantly accelerates computationally intensive simulations of plasma in fusion reactors. By using advanced machine learning surrogate models, researchers can more quickly analyze various operating conditions and deepen their understanding of physical processes, while maintaining high accuracy in their results. Such approaches not only contribute to scientific breakthroughs but also create new opportunities for involving students in research and advancing knowledge within the university environment.

Advances in AI are increasingly influencing various areas of society and bringing about changes in the workplace and in everyday life. The development of technologies such as humanoid robots shows that technological progress often outpaces the adaptation of social and ethical frameworks. Such discrepancies further underscore the importance of education as a space for understanding and critically examining technological changes. Universities play a vital role in preparing individuals for life and work in a rapidly changing environment. A critical understanding of technology is thus becoming one of the key tasks of contemporary higher education.

At the level of education and broader digital policies, coordinated frameworks for the development and use of AI are increasingly taking shape. European initiatives emphasize the strategic importance of developing indigenous technologies and strengthening research and digital competencies as the foundation for European technological sovereignty. At the same time, the standardization of rules and guidelines for the use of AI is becoming increasingly prevalent in higher education, as institutions are gradually shifting from ad hoc decisions to more systematic and comparable approaches. The goal of these efforts is to ensure the safe, ethical, and effective use of technology, as well as greater transparency regarding accountability in its use. At the European level, there are also initiatives to accelerate the adoption of AI in various sectors, seeking a balance between regulation and innovation. In Slovenia, events such as the Slovenian Informatics Days illustrate concrete steps toward achieving digital transformation and highlight the importance of cooperation between the government, the business sector, and the research and education communities.

Discussions in Slovenia are increasingly based on specific research examples of AI applications. The program Ultrazvok features the development of ARIA, a Slovenian artificial intelligence-based radiology assistant, which was developed by a research group led by Prof. Dr. Janez Žibert at the University of Ljubljana’s Faculty of Health Sciences. The system assists physicians in analyzing magnetic resonance images of the prostate and lungs and serves as a decision-making aid, not as a substitute for professional judgment. The discussion emphasizes the importance of understanding how such systems work and their limitations, as well as a clear division of responsibility between humans and technology. The ARIA case thus demonstrates how AI can contribute to the advancement of knowledge and the improvement of professional practices within the Slovenian university environment.

The articles in this newsletter show that the issue of AI in higher education is no longer a topic for the future, but one of the present. The shift from thoughtful pedagogical approaches to systemic policies increasingly confirms that the effective use of AI requires an understanding of context, the cooperation of the academic community, and clearly defined values of university education.

If you have explored an interesting use of AI in education at your faculty, in a course, or as part of your professional work, we cordially invite you to share your example with us via the form for collecting examples of best practices. We look forward to reviewing the submissions and including selected contributions in future issues of the AI newsletter so that, together, we can build an overview of useful and thoughtful approaches in the university setting.

Authors: Mateja Bevčič, Eva Kern Nanut, Eva Škraba, Maja Kosmač, Sanja Jedrinović Čufer, University of Ljubljana Center for the use of ICT in pedagogical process