UVA Today

UVA Today, Supplemental, Subscribe

• Faculty Spotlight
• Hoo? What? Where?
• Inside UVA Podcast
• Made for the USA
• The Music Beat
• So Hoos Asking?
• UVA In The News
• Video

Q&A: In the age of AI, what is a library for?

• By Alice Berry, aberry@virginia.edu, April 15, 2026

In December 2022, when University of Virginia Librarian and Dean of Libraries Leo Lo worked for another institution, students began asking for books and articles that did not exist. It alarmed him.

ChatGPT, OpenAI’s consumer-facing artificial intelligence chatbot, had been available for a few weeks. The authors and the titles of the research seemed plausible, but were complete fabrications.

“Those students needed a librarian before they needed a better AI tool,” Lo said.

As the leader of UVA Library, Lo, along with Christa Acampora, dean of the College and Graduate School of Arts & Sciences, plans to launch an “AI Literacy and Action Lab” Friday. The goal is to teach students how to use AI to learn literacy and critical thinking skills. 

Libraries are in a position to help train AI models, too. Lo has initiated a statement of shared practice for UVA Library and partner institutions, including Duke University, Oklahoma State University and Northwestern University. AI companies are increasingly reliant on archival materials held in libraries to train their models as they work through open material on the internet, and the statement provides a framework for responding to AI companies’ requests.

UVA Today talked to Lo about what a library does in the age of AI.

Q. Where do you fall on the AI enthusiast to AI detractor spectrum?

A. A faculty member at another university asked me recently whether it was defensible to ban AI in her course. I said yes.

That probably isn’t what people expect from someone who spent the last three years building a framework for AI literacy. But it was the honest answer for now. She believed her students needed to develop a specific skill that AI use would short-circuit, and banning it was the right call for that course.

What I would ask of faculty who choose that path is to stay open, keep up with how the technology is developing, and be willing to try approaches others have tested. That is part of what the lab is for: to produce case studies that give faculty something real to work from when they are ready to revisit the question.

I’m wary of the two confident positions on AI in higher education right now: the people certain it will transform teaching, and the people certain it will destroy it. Both are getting ahead of what we actually know about what’s happening in our classrooms.

Q. What is the function of a library in this AI age?

A. A research library has always done two things: help people find information, and help them judge it. AI changes the tools, not the mission. If anything, the mission gets sharper. The library is also one of the few places in a university built to convene across disciplines, and AI literacy requires exactly that: technical knowledge, ethics, critical thinking, practical skill, and societal impact all at once. No single department owns that combination. 

A library can hold it together. That is why we are launching the AI Literacy and Action Lab here. Dean Acampora and I share the conviction that AI is an opportunity for the liberal arts, not a threat to them. The lab is built on that shared premise: AI literacy is a liberal arts problem as much as a technical one, and a university that treats it only as technical will get the answer wrong.

Q. Does AI make information more accessible?

A. Yes, but I want to be careful about what the word “accessible” is doing in that sentence.

AI makes information easier to retrieve and easier to draft into a first version of almost anything. But access to information is not the same as access to understanding. A fluent summary from a language model can give someone the feeling of having learned something, without the substance of it. The lab is designed around that distinction. We use what happens in real classrooms as data, working with faculty and students to develop ways to teach and learn with AI that build understanding, not just output.

Q. What are the pitfalls of more accessible information?

A. We have moved, almost without noticing, into what I call “an AI answer economy.” For most of the history of research, finding information meant navigating sources: chasing a citation, comparing two accounts, deciding which author to trust. AI collapses that work into a single synthesized answer.

That shift sounds like progress, and in some ways it is. But it quietly relocates the hard part of learning. When the answer comes first, the verification step has to be added back in deliberately, or it disappears. Students stop opening sources. Researchers stop comparing claims. Trust moves from where an idea came from to how confidently it is phrased. An answer economy without strong habits of verification is not a more informed society. It is a faster one, producing graduates who are quicker, but not wiser.

Continue/Read Original Aricle: Q&A: In the age of AI, what is a library for?