AI in Academic Libraries: Success, Pitfalls, Perceptions, and Why We Need AI Literacy

AI in Academic Libraries: Success, Pitfalls, Perceptions, and Why We Need AI Literacy

Copyright: © 2024 |Pages: 26
DOI: 10.4018/979-8-3693-1573-6.ch002
OnDemand:
(Individual Chapters)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

This chapter discusses the implementation of artificial intelligence (AI) in academic libraries and the impact of AI related technologies on various library services. AI impacts research and educational activities in academic institutions so they can provide adequate operations and collaborative technologies in order to support diverse curricula and research. The complexity of research workflows and their variability couples with increasing demand for effective personalized education, so it acts as a driving force for application of AI related technologies. AI assisted library services demand reimaging traditional libraries' roles, so that librarians would become not only users of AI technologies but also co-creators and experts capable of conceptualizing AI driven services to library patrons. That requires librarians to promote AI literacy as a framework encompassing research literacy and data literacy, and understood as a necessary skill set for effectively communicating in an AI driven medium.
Chapter Preview
Top

Definitions And Types Of Ai

The term “artificial intelligence” (AI) is both an opaque and yet powerful expression which startles many librarians. The definition given by John McCarthy in 1956, at the first ever AI conference, describes AI as “the science and engineering of making intelligent machines”. That characterization, however, diverges significantly from modern understanding of the entity of AI. The confusion comes from the fact that the term has at the same time scientific, technical, and social designations. Consequently, there are formal and informal AI definitions which exist in parallel. The informal definitions are associated with humanistic perspective of AI and thus typically counterpoint machine-based AI to native human intelligence. This designation seeks to answer the fundamental question “Can machines think?”, formulated in the late 1950s by Alan Turing. He invented a test named after him (Turing test) to determine machine’s “intelligent behavior” by looking at whether or not “the system acts like a human”. Per this test the intelligence assessment of the machine is based on the computer’s ability to produce responses indistinguishable from the ones generated by a human in a course of human-computer interaction. The humanistic approach focuses on generated knowledge concerning the interaction between machine and human intelligence including technologies. This view caused some researchers to introduce other definitions of AI inspired by different characteristics of human and machine rooted intelligence. Thus, De Cremer and Kasparov (2021) juxtapose machine associated AI to Authentic (human) Intelligence and define a third type of AI –Augmented AI (AuI). The latter is a machine-complemented Authentic Intelligence in which the machine deals with voluminous data tasks and calculations, but the human takes a leading role in switching from short term to long term strategies using his/her ability to anticipate, feel, and judge. It is important to mention that De Cremer and Kasparov drive the distinguished line between machine intelligence and human intelligence based on types of tasks suitable for one or the other. Thus, machine AI is understood only as useful for closed management systems with well-defined scenarios while the openness was attributed only to human rooted intelligence.

Key Terms in this Chapter

AI Literacy: AI literacy cannot be precisely defined and should be explained as a broad convolution of: knowledge about AI principles, ability to use AI in tools, services and applications, and ability to critically evaluate AI output.

Generative AI: A set of ML models that can generate new data, i.e., the output is object that look like the one on which model is trained; ChatGPT is the most popular instance.

Artificial Intelligence in Libraries: That is a synergy between computing and psychology used to increase librarians’ productivity, quality of services; quality of information yields and quality of communication with patrons.

Machine Learning: Machine Learning (ML) allows for the programming of the computers by using data instead of explicit instructions.

Artificial Intelligence: Artificial intelligence (AI) is defined as intelligent performance expressed by artificial body usually a computer program.

Complete Chapter List

Search this Book:
Reset