This library guide is meant to help search for information regarding artificial intelligence (AI), and to introduce to you to how AI tools are being used for reference and research.
The guide is not designed to address how AI tools can be used as writing or proofreading aids, and all students should read their course syllabi and discuss with their professors before using AI in coursework. It also does not address how AI tools may be used in pedagogy. Please see the disclaimer at the bottom of this page for additional information.
This library guide is still being developed, so stay tuned for more! Pages that are still in development include:
|
Libraries are places that both store and offer access to enormous amounts of information, but many people do not know that librarians, archivists, catalogers, and other library staff are known as "information professionals." In addition to academic databases and the books and articles within, information exists in many forms outside the walls of the library. Thus, depending on your need, librarians may be able to help you:
Academic databases are information systems that are designed to provide you with results in response to the inputs ("queries") you put into them. Meanwhile, while they look similar on the surface, search engines like Google are constructed in a fundamentally different way to meet different needs. When applications of AI are embedded into these systems, they can become more complex and powerful, but that also means that users have to overcome larger learning curves to effectively find and access the information they need.
While algorithmic search optimization has been a feature of most academic databases for a very long time, AI-research assistant features are not yet turned on in our databases or catalog. If new AI-driven features are enabled in any of our library databases or catalog, we will work to include information and tutorials for their use.
Note that AI research assistants that are being developed for academic databases are for the most part using Retrieval-Augmented Generation (RAG): this means that even though the Large Language Model (LLM) that allows the tool to understand English was trained on large bodies of text like the internet, the information that the tool retrieves comes from set of data. For example, a research assistant for a library catalog using RAG could answer your questions using only the information available in the catalog, not from Google.
Even though the features are not currently turned on in our databases, you can explore what academic database vendors are saying about AI research assistants below:
Because artificial intelligence has huge impacts on the use, distribution, adaptation, creation, and discovery of information, it is important for librarians to understand it and to offer assistance in navigating an AI-driven world.
Artificial intelligence is defined in the Minnesota State "Generative Artificial Intelligence" guidance document as "a branch of computer science devoted to developing data processing systems that perform functions normally associated with human intelligence, such as reasoning, learning, and self-improvement" (2024, p.1). It has existed in many forms for decades, but its impact has reached new heights recently with the development and popularization of more generative AI tools. Together, we can take on this new big learning curve, and hopefully find the resources necessary to critically evaluate AI, and to use those tools responsibly if you so choose.
Forms of artificial intelligence that you may interact with in your daily life, but may never have considered to be "AI" include:
This means you are also engaging with products of "machine learning" daily. Machine learning is a subcategory of artificial intelligence where machines are taught by being fed data. They then find patterns within that data, and attempt to adapt based on the patterns they find. This is different from traditional programming, where machines are taught to follow specific rules.
So, for example, consider this:
A simple autocorrect may only refer to a core dictionary and some rules written by the programmer. This is AI, but not necessarily machine learning. Meanwhile, if your predictive text and autocorrect in your texting app adapts based on your unique typing style and use of language, it is likely an example of machine learning.