The 2nd Digital Transformation Day, entitled Responsible Use of Artificial Intelligence, was held at the University of Zagreb on Friday, 23 October.
The event focused on the challenges and opportunities of using artificial intelligence in higher education, science and research. It brought together representatives of the academic community and experts who discussed the impact of artificial intelligence on the experience of studying and conducting research, as well as on the university's mission and research culture. Participants also discussed the challenges posed by the use of artificial intelligence and its responsible application in educational and research contexts.
Following the keynote lecture delivered by Professor Luciano Saso from Sapienza University of Rome, who is also Secretary General of the UNICA network of universities from the capitals of Europe, a panel discussion was held on the responsible use of artificial intelligence in higher education and science, with an emphasis on its systematic integration into university processes. The panellists were Associate Professor Dijana Oreški, Ph.D., from the Faculty of Organization and Informatics; Associate Professor Marko Horvat, Ph.D., from the Faculty of Electrical Engineering and Computing; Assistant Professor Melita Milenković, Ph.D., from the Faculty of Transport and Traffic Sciences; Juraj Bilić from the Croatian Academic and Research Network CARNET; and Ivan Marić, Director of the University Computing Centre SRCE.
The panellists discussed the systematic introduction of artificial intelligence into the educational process. They also addressed the application of the European legal and regulatory framework, the importance of managing institutional data assets, the development of domain-specific and open-weight language models, and the redefinition of the roles of teachers and students.
Speaking about the importance of protecting and using institutions' own data, Ivan Marić, Director of SRCE, emphasised that the greatest asset of institutions lies in their data assets, which comprise various digital objects stored in repositories in Dabar, scholarly and professional papers in Hrčak, and e-courses on Merlin. He added that the future lies in developing specialised, domain-specific artificial intelligence tools trained on this scientifically validated data.
