Cataloguing is one of the areas of work to which we allocate the most librarian staff at NUK. The work is requires professional expertise and is time-consuming, librarians are already overworked in the current situation, and in the era of generative artificial intelligence, an exponential growth in the amount of material to be processed is not excluded. Therefore, it is crucial to enable employees to use technical solutions that make their work easier and save time. At NUK, we focus in this area on the use of solutions supported by open large language models, where we prioritize semi-automation of routine work that does not require professional insight. When evaluating potential technical solutions, several factors are important, among which an optimal balance must be found. Usually, the quality of results and the speed of obtaining answers are in an inversely proportional relationship, together with costs; it is important to take into account the capabilities of the hardware and software available to us; development must be coordinated with the actual work process and the needs of cataloguers; last but not least, compliance with laws must be ensured, where copyright protection is of key importance. You will find out how we deal with these issues at NUK in this presentation.
