Generative artificial intelligence (GAI) is a form of AI designed to generate new content – text, images, audio, or video – using large language models (LLMs). In recent years, a range of LLM-based tools, such as ChatGPT and Gemini, have been released and have rapidly gained widespread popularity among users. Today, GAI’s potential is already being used across many areas of everyday life, and all signs suggest that this form of artificial intelligence will play a key role in shaping their future development.
One field that can benefit significantly from generative AI is the academic library. On the one hand, GAI can support librarians in their daily work; on the other, it can improve users’ access to library resources and enhance their availability. But are Polish university libraries ready for the revolution that artificial intelligence is bringing? This question was explored by Prof. Grzegorz Gmiterek from the Faculty of Journalism, Information, and Book Studies at the University of Warsaw, and Dr. Sebastian D. Kotuła from the Department of Information and Digital Culture at Maria Curie-Skłodowska University in Lublin.
How can GAI help libraries?
So how exactly can generative AI support librarians and users of academic libraries? One of its key strengths lies in improving information retrieval through advanced natural language processing tools and specialized recommendation systems. It can also significantly speed up this process thanks to its ability to analyze and process large datasets quickly and efficiently.
Generative AI also enables the automated summarization of sources and publications, helping users quickly determine their relevance. Additionally, artificial intelligence is widely used in optical character recognition (OCR), enabling the conversion of printed or handwritten text into digital form.
However, the potential of GAI goes even further. It can enhance library accessibility through automated chatbots and virtual assistants that respond to frequently asked questions, allowing users to resolve some issues even when library staff are not available. Moreover, such systems can offer personalized recommendations tailored to individual users, making library resources easier to navigate and more effectively used.
Ready for a revolution?
Prof. Gmiterek and Dr. Kotuła set out to examine how academic libraries in Poland are responding to the challenges posed by the rapid development of generative artificial intelligence. Their study consisted of two stages. The first was an anonymous survey conducted in July 2024, sent to the directors of 65 Polish academic libraries. The researchers aimed, among other things, to find out whether these libraries are already using GAI tools, whether they plan to introduce them, and whether current working conditions are suitable for their implementation. In total, 28 institutions took part in this phase of the study, representing 43% of all university libraries in Poland (according to the list of public universities compiled by the Minister of Science and Higher Education in 2024).
The second stage involved an analysis of the websites and social media profiles of all 65 institutions, carried out in October 2024. The researchers examined materials published in 2023 and 2024, identifying 118 social media accounts across platforms such as Facebook, Instagram, X (Twitter), YouTube, TikTok, and Pinterest. This content analysis aimed to determine whether academic libraries communicate about GAI and AI online, and if so, what kinds of messages they share.
In the survey, 46.4% of respondents reported that their libraries use generative AI tools to some extent, while 42.9% indicated that they do not use them. Meanwhile, 39.3% said their institutions are interested in topics related to generative AI, and nearly one-third of this group noted that staff are encouraged to explore its potential. However, only 7.1% of respondents reported that formal regulations or guidelines have been introduced to support the use of such tools. Overall, 57% of participants supported the implementation of GAI in their libraries. At the same time, respondents pointed to a lack of expertise among librarians and the absence of clear regulations as the main barriers to adoption.
Challenges of the new era
What conclusions did the researchers draw? Above all, they identified a clear and growing interest among Polish academic libraries in the opportunities offered by generative artificial intelligence. This is evident not only in the survey results but also in the analysis of content these institutions publish online. In 2023 and 2024, libraries increasingly shared materials on this topic; however, their posts did not consistently distinguish between generative artificial intelligence and artificial intelligence in general. According to the researchers, this distinction was not considered particularly important by these institutions, which – together with the survey findings – may point to a limited understanding of the differences between GAI and artificial intelligence in general.
At the same time, Polish academic libraries face a range of challenges surrounding the implementation of generative artificial intelligence. As Prof. Gmiterek explains:
“The first issue, of course, is funding and the cost of implementation, as well as training for staff and students. Most students aren’t even aware of the opportunities libraries already offer when it comes to tools that use generative AI – and that awareness is crucial. Second, there’s the challenge of integrating AI tools with existing infrastructure, like libraries and digital repositories. That kind of integration opens up new possibilities for users, especially what we call ‘conversations with documents.’ Third, libraries need specialists who can handle a wide range of user questions – both about what these tools can do and where their limits are. These experts are starting to emerge, but there are still too few of them. In the West, they’re referred to as ‘AI Librarians.’ They’re the ones who bridge the gap between AI tools and reliable, verified information,” the researcher explains.
Noteworthy, according to the study participants, generative AI tools are already being used in some academic libraries in Poland. Prof. Gmiterek also refers to further research conducted jointly with Dr. Kotuła:
“In March and April 2025, we conducted a survey on libraries’ use of the AI tool Primo Research Assistant (beta version). The article based on this research is now finished, and we’ve submitted it to an international journal. The article has been accepted for publication. What makes this topic particularly interesting is that, in theory, this tool has the potential to be widely adopted by Polish libraries – not only academic ones. The Assistant is a feature integrated into the Primo academic search engine. It allows users to search multiple information sources at the same time, such as databases, libraries, and digital repositories. It also includes a generative AI component that provides answers to users’ questions. Some academic libraries have already decided to introduce this feature – for example, Maria Curie-Skłodowska University in Lublin. Their experiences are definitely worth following and analyzing, and that’s exactly what we’ll be doing in the coming months,” Prof. Gmiterek concludes.
The text was originally published in Polish on the Serwis Naukowy UW website on June 18, 2025. It was updated in May 2026.
