Artificial Intelligence

Artificial intelligence (AI), in many forms, has long been a feature of research and practice in information science. Defined broadly as computers that “learn” from environmental inputs and adjust outputs accordingly, AI has a long history in information search systems, for instance. The more recent iterations of AI, namely large language models (LLMs) and generative tools such as ChatGPT, have caught the world’s attention, and are the subject of significant discussion in multiple contexts, including higher education. This panel focuses on the implications of these tools for information schools. One obvious implication is the need to expand the curriculum, at least at the master’s level, to ensure that graduates are familiar with these technologies, their affordances, and their challenges. Pedagogical implications include concerns for how best to teach these topics, as well as how to manage students’ use of LLMs in their research and writing. Multiple curricular and pedagogical issues merit discussion, and implications for programmatic or institutional policies should be included in those conversations. The panel will begin with a brief introduction by the moderator (Heidi Julien), followed by 10-minute presentations by three scholars with expertise and experience in these areas (Sam Abramovich, Sean Burns, and Saguna Shankar). A 45-minute opportunity for audience discussion will follow, in small groups first if audience size allows, followed by a 15-minute wrap-up session to summarize outstanding concerns and issues, as well as to identify opportunities for collaboration to address these matters. Dr. Sam Abramovich is an Associate Professor (Departments of Information Science and of Learning and Instruction) and Associate Dean for Academic Affairs, Graduate School of Education, at the University at Buffalo. In his presentation, he will share insights gained from teaching an online course for Master of Library Science students focused on emerging technologies. Over multiple iterations of this course, he observed that many students were quietly using generative AI tools, like ChatGPT, to generate responses for assignments and discussion board posts. This practice undermined the course’s learning objectives, so he redesigned the assessments to explicitly allow and encourage AI usage, as long as students acknowledged how they employed these technologies. Surprisingly, almost no one admitted to using AI, even though his own analysis strongly suggested that many did. This lack of transparency is indicative of a larger culture of fear and secrecy surrounding AI use in Information Science education. Yet, these tools are increasingly pervasive, nearly impossible to detect, and are quickly becoming integral to library work. In his presentation, Dr. Abramovich will argue that Information Science programs must immediately embrace and normalize the use of generative AI. Otherwise, the Master of Library Science degree risks losing its relevance, with its demise hidden until the impact becomes evident and irreparable in the coming years. Therefore, failing to adapt could undermine our field’s foundation. However, by embracing AI into our curricula, we can empower future librarians to thrive in a rapidly changing information landscape.; Dr. C. Sean Burns is an Associate Professor in the School of Information Science at the University of Kentucky. He will focus on “Shortcuts, Learning, and the Sacredness of Writing.” Our culture sends mixed messages, stressing the importance of learning while rewarding the product, rather than the process. But learning is a process and not a product. When students use LLMs to take shortcuts in assignments, they reflect a deeper societal tension that prioritizes extrinsic outcomes over intrinsic learning. This raises the question: what do we truly value, the product or the process? If it is the product, then taking shortcuts is entirely rational. We send other mixed messages. We say that writing is thinking (Zinsser, 1988), but we think in other ways, too. Transcribing interviews fosters deep intimacy with the data. Manually calculating sums of squares for an ANOVA builds understanding of statistical reasoning. In these and other cases, this engagement transforms us into scholars and not just researchers. Yet, we have handed over both tasks to computers. Why, then, are some forms of thinking, like writing, held sacred while others are not? What rule determines which shortcuts are acceptable and which are not? Dr. Burns will explore this tension within the context of generative AI tools like LLMs. Drawing on his research on ungrading and Universal Design for Learning (UDL), he will challenge assumptions about assessment, learning, and technology’s role in shaping how we engage with knowledge. Dr. Saguna Shankar is an Assistant Professor in the Department of Information Science at the University at Buffalo. She will discuss “Practice and Pluralism in Curricular Considerations”. While media headlines on AI often claim unprecedented innovations, existing expertise in library, archival, and information science receives less attention. Within the field, there is agreement about the critical roles and responsibilities of information professionals in shaping ethical considerations and regulations of automated, algorithmic, and AI-based systems. Threading these roles and responsibilities through curricula is a priority for educators and practitioners. Dr. Shankar will reflect on emerging professionals’ sensemaking about AI from her experiences teaching information policy. Challenges in teaching information policy, however, include addressing narratives of newness, hype, and fear, interrogating inevitability, and an erosion of agency prompted by contemporary overreach in the economics and politics of the technology industry. Drawing on her research, she will also share related issues for leveraging existing expertise and amplifying opportunities for information professionals’ participation in policy analysis and sociotechnical decision making.