Module 1: Connecting Libraries to the AI Ecosystem
AI Literacy Essentials for Academic Libraries: Beyond the Basics
Choice and Clarivate have teamed up once again to expand last year’s eight-week, newsletter-based course on generative AI literacy for academic library workers by creating three new modules. Each module contains bite-sized, self-paced material such as readings, multimedia content, case studies, and interviews with thought leaders.
Make sure you’re registered to receive all three modules in your inbox! If you missed the original eight-part series, you can register here.
In this module, we look at connecting libraries to the AI ecosystem.
Learning Objectives:
- Navigating ethical concerns with AI while staying informed on new technologies
- Outlining current library content AI integration strategies
- Tips for benchmarking and evaluating research AI tools/layers
- Understanding the library’s role and value in the AI ecosystem on campus
- Discovering how students, faculty, and researchers are responding to AI in their scholarship
Table of Contents
Introduction
Academic Libraries at the Center of Campus AI Strategy
How libraries are leading discussions on connecting AI tools to scholarly content
Right now, groups across your campus are discussing artificial intelligence. Where is your library in these conversations? Central… or peripheral? Leaders of AI policy… or downstream implementers? What role should the library play in campus AI strategies?
As an academic librarian who has—at times strategically, at times serendipitously—been included in many of these conversations, my answer is simple: Every serious conversation about AI, at every one of our institutions, needs libraries at the table.
To our collective detriment, this isn’t the default. All too often, librarians are brought in late or are asked to carry out decisions that have already been made. At times, perhaps due to institutional culture or professional humility, we inadvertently sideline ourselves. This is not a reflection of our value; it is a reflection of how narrowly AI leadership is often defined.
Libraries have so much to contribute to these conversations, including relevant skills, institutional trust, and a long history of operationalizing our values amidst enormous technological change. At a moment when many campuses are asking what “responsible AI” looks like in practice, libraries have hard-won expertise in data privacy, equitable access, addressing bias, rights management, and helping users navigate complex information environments. Our existing frameworks of evaluating information through a lens of authority, context, process, and systems are more relevant than ever in an AI-mediated world.
Connecting AI Tools to Scholarly Content
This experience can be applied to evaluating, selecting, and licensing AI tools through a human-centered framework that considers the questions such as: What does this deployment of AI do to and for the user? Does this tool or AI deployment help fulfill our mission of advancing knowledge and discovery? Does it have a place in scholarship offered through the library? The criteria by which libraries implement AI offerings into their resources can act as a model for the institution in terms of AI use, responsibility, and research practices. The tools libraries choose to evaluate, license, build, and promote will increasingly shape how students discover sources, how faculty conduct research, and how scholarly knowledge is incorporated into teaching and learning. Libraries can establish expectations for transparency and attribution and evaluate emerging tools before they become embedded in academic workflows to ensure that scholarly values such as reproducibility and rigor remain visible as AI increasingly mediates access to information.
There is also a growing recognition that academic libraries are becoming platforms through which AI interacts with scholarly knowledge. Vendors are embedding AI-powered research assistants into discovery systems and databases. Universities are experimenting with connecting large language models (LLMs) to institutional repositories, archival collections, and licensed scholarly content. In my digital humanities lab at the University of Utah, for example, we recently created an AI-powered chatbot that will allow users to interact conversationally with our historical newspaper collection. Increasingly, the flow is also moving in the opposite direction. Rather than beginning with library systems, vendors and publishers are exploring ways to connect scholarly content directly into general-purpose AI assistants, allowing users to discover and interact with library resources from within the AI tools they already use. As these tools evolve, decisions about what information AI systems can access, how they retrieve it, how they cite it, and how they represent uncertainty become profoundly important. These are questions libraries have been addressing for decades in other forms.
Opportunities for the Library to Lead AI-Enabled Discovery
The library’s collections are particularly valuable in this environment and hold incredible promise for AI-enabled discovery. As Dan Cohen and others have observed, continued advances in AI models rely on more and better data, especially as the current models have exhausted what is available online. The Public Interest Corpus represents one recent effort to proactively and ethically make library collections available as training data for AI developers and scholars alike to both improve commercial models and unlock the promises of AI research and discovery.
LLMs trained on web-scale data have demonstrated impressive capabilities, but many of society’s most important challenges will require access to high-quality, domain-specific, and curated data. As humanity’s most comprehensive and well-edited repositories of human knowledge, libraries have the opportunity to negotiate wisely in this moment by democratizing access to training data, powering new forms of research and AI-assisted discovery, shaping ethical and legal frameworks for AI development, and ensuring that the next generation of AI tools reflects scholarly values rather than solely commercial interests (Google DeepMind’s AlphaFold stands out as a prime example of utilizing well-structured, openly available training data for a scientific breakthrough).
AI has created a tremendous opportunity for libraries to lead. San José State University’s Dr. Martin Luther King, Jr. Library has positioned itself at the center of the institution’s AI strategy through the creation of an AI Center for Civic and Social Good, developed in partnership with the City of San José and Adobe. Rather than waiting to support AI initiatives designed elsewhere, the King Library has become a catalyst, convener, builder, and strategic partner in the campus AI ecosystem.
Claiming a Seat at the Table
At my own institution, I served as co-director of the Summer Institute for Higher Education Faculty on “Humanities Perspectives on Artificial Intelligence” and have become a frequent speaker and collaborator on the social and ethical implications of AI. Serving as a panelist alongside leaders from industry, government, and academia, I surface concerns around labor, surveillance, bias, misinformation, environmental impacts, and cognitive decline, not as an AI-skeptic per se, but as someone who believes we have the agency to build human-centered AI (or something closer to it) via thoughtful decision-making and good governance. In these conversations, I am continually struck by how hungry people are for technology that serves rather than exploits human beings. They want systems that advance learning, creativity, and connection—not simply efficiency or profit. Libraries are uniquely positioned to help articulate and build a vision where we don’t “move fast and break things”—especially when what’s being broken are people, our social fabric, and the natural world we depend on—and one where we don’t treat privacy like collateral damage in exchange for the convenience these tools afford us.
If libraries are to inhabit this role, however, it will not happen automatically. We must be intentional about claiming our place in these conversations and we need to do it now, while decisions are still being made. Sitting at the center of campus AI strategy means setting aside imposter syndrome (if not for personal ambition than for the good of the order), building relationships, showing up—visibly and consistently—where decisions are being made, and moving beyond critique to demonstrate what responsible AI looks like in practice.
As AI is increasingly embedded in research, teaching, publishing, and discovery, academic libraries have an opportunity to become indispensable resources and guides. We can help shape how AI systems access, interpret, and connect scholarly knowledge across our institutions. This benefits our users through better research workflows, stronger information literacy, and more trustworthy systems. It also positions libraries as central partners in the university’s AI future rather than service providers responding to decisions made elsewhere. The future of AI in higher education is being written now. Libraries have both the expertise and the responsibility to help write it.
Roundtable Discussion
Connecting AI to Scholarly Content through the Library: Challenges, Opportunities, and Implications
We explore how AI deployment across an institution can make the academic library central to AI strategy, and the implications of connecting AI to scholarly content via the library.
Expect to learn about the AI layers and research assistants libraries are currently navigating, strategies to benchmark AI products for library content, and how students and faculty are adjusting research workflows in response to AI. Further, our panelists chat about encouraging robust research skills amidst an influx of AI use and the balance struck between enforcing a library’s mission and staying up-to-date on how new technologies will impact patrons.
This discussion features:
- Maxwell Gray, Digital Scholarship Librarian in Raynor Library, Marquette University
- Elizabeth Szkirpan, Library and Information Studies Professional, Harvard Business School’s Baker Library
- Evan Simpson, Associate Dean for Experiential Learning and Academic Engagement, Northeastern University
- Miri Botzer, VP Product Innovation, Clarivate
Reading Room
A selection of multimedia content to fit your time commitment.
5 minutes
- “Yale Library is testing an AI-powered connection for deeper searching of library resources.” Yale Library, 2026.
- “Benchmarking AI Models in the Library: Critical Questions for Library Professionals and Users.” LibTech Insights, 2025.
- “Why 95% Of AI Pilots Fail, And What Business Leaders Should Do Instead.” Forbes, 2025.
10 minutes
- “Adoption of artificial intelligence in academic libraries: A systematic review of current practices, challenges, and research opportunities.” The Journal of Academic Librarianship, 2026.
- “Student Perspectives on Using Generative Artificial Intelligence for Research: A Qualitative Approach.” portal: Libraries and the Academy, 2025.
- AI Tools for Academic Libraries: AI Research Assistants. LibTech Insights, 2025.
20+ minutes
- “A Landscape of Artificial Intelligence in Libraries, or ‘What Librarians Talk About When They Talk About AI.'” CNI Fall 2025 Membership Meeting. Video.
- “Claude AI and Literature Reviews: An Experiment in Utility and Ethical Use.” Library Trends, 2025.
- “Discovery and Use Reimagined: Connecting Scholarly Collections and Artificial Intelligence Workflows.” CNI Fall 2025 Membership Meeting. Video.
Survey
In this brief, seven-question survey, we want to learn more about your library’s approach to connecting AI to library content and AI use on campus.
Discussion Questions
AI gives us a lot to think about. Share your thoughts in the Leave a Reply section below!
- How is your institution tracking student engagement with scholarly content through AI tools?
- What measures does your library use to evaluate GenAI models for library collections?
Coming Up
Next month we’ll explore the pedagogical road map for teaching AI literacy. The module will include an introductory essay, activity bank provided by academic librarians, reading list, and more!
Take me to the Table of Contents.
Know someone who might be interested in this series? Share the registration page with your colleagues!
Make sure you’re registered to receive all three modules directly in your inbox.
Learn more about Clarivate and Choice’s LibTech Insights.
53 responses to “Module 1: Connecting Libraries to the AI Ecosystem ”
-
How is your institution tracking student engagement with scholarly content through AI tools? We don’t. We are still struggling with developing a clear vision regarding AI.
What measures does your library use to evaluate GenAI models for library collections? We currently don’t. -
We are not currently tracking student engagement with AI tools. However, I do introduce students to the AI tools integrated into our library databases.
The college uses Turnitin to check for plagiarism and also utilizes its AI detection capabilities as part of our academic integrity efforts. Faculty approaches to AI use vary, with some instructors permitting AI tools and others not allowing their use.
At this time, the library does not formally evaluate generative AI (GenAI) models for inclusion in library collections, nor does the library currently have a formal policy governing the use of AI.
-
I am unaware of any tracking of student engagement through AI tools at our institution. The institution has approved AI tools for faculty, staff, and student use. The library was not a part of any decisions on what AI tools to use. There is a library AI working group being formed, which may address evaluating GenAI tools for collections.
-
We currently do not track.
-
As a small community college, we currently do not track AI usage by students (to my knowledge). Our library also does not evaluate genAI models in library resources at this time, as we have three librarians between two locations. We have familiarized ourselves with various AI products, but have not considered their usage in relationship to the library collection itself.
-
We are not tracking student engagement with AI tools. But we form library circles, which is a small group session inviting students to meet with the librarians and let us understand their needs, library workshops on AI and exhibitions or AI tools. We ask about their habits on using AI tools in these activities and irregular survey. We have a team dealing with database subscription, also handling the AI tools selection as these tools usually in relation with the databases.
-
How is your institution tracking student engagement with scholarly content through AI tools? A curriculum and instruction committee identified all courses with AI focus or discoverable AI-related content. There was a campus-wide survey done last year.
What measures does your library use to evaluate GenAI models for library collections? We haven’t done any formal evaluations as of yet, but we have an AI task force in place and just hired an AI/Emerging Technologies Librarian that will be involved in evaluating AI models for library use. -
My institution is not currently (nor do I think we plan to) track student engagement with AI. We also do not, to my knowledge, use GenAI for the library collection.
The only policy my institution has about AI is on whether or not students may use AI for courses. -
We do not do any of these processes as our institution is still defining and debating the AI in academics’ question. Personally, I think that tracking student usage of research content through AI tools is problematic both from a technical and ethical standpoint.
-
As our small community college does not have an AI policy in place, the library does not track any student engagement with AI. It is up to the faculty to decide whether or not to allow the use of AI & which AI programs may be used. The library staff refer the students to their individual faculty. The library staff explain the use of AI to patrons covering, & verification of results, accuracy, plagiarism, copyright, protection of confidential information. Most faculty have asked that we disable any AI additions to our vendor software.
-
How is your institution tracking student engagement with scholarly content through AI tools?
We have been using the engagement information that is included in a few of the database/platform reports. From the reports, the numbers show that the students are not using them. -
We do not officially track genAI use yet. However, I do research on the topic: do university students use genAI to find info for tour assignments. I did survey and interviews. I was surprised that out of the 48 participants in ht interviews, only 14 used AI at any level to find info for research assignments. The majority used the library databases and were happy with what they found through traditional searching. They are reluctant to use genAI because they see it as a potential crutch, detrimental to learning, and brought with ethical problems, especially as it pertains the environmental cost.
-
Our institution has a list of approved AI tools. However, I am not aware of the institution tracking student engagement with scholarly content through AI tools.
The library is forming a committee on AI, which I expect will address the evaluation of GenAI models. -
At this time, we are not formally tracking student engagement with scholarly content using AI tools. We are gauging their usage informally by asking them in IL sessions and workshops how they are utilizing different AI tools.
I think this is a time of exploration and thoughtfulness for a lot of our librarians. Reading about other library workers’ frameworks, benchmarks, and perspectives have been helpful in our conversations to formulate our own. Again, nothing formally, though, as we also await institutional policies and guidance from our college.
-
My institution doesn’t have artificial intelligence as a library tool, except for the one integrated into Google. We don’t yet have a program to measure its use. Students are guided on its use through workshops, but it’s only basic training. We understand that this data is being lost, since students do use AI for academic work, but the library staff is concerned that this use might not be correct or ethical.
-
As far as I’m aware, our institution doesn’t actively track student usage of outside models. The only internal model we have is a helper chatbot that is a generic, campus wide focus. It’s been trained on our LibGuides and internal “knowledge base” but doesn’t have any access to library collections.
The only library focused AI tools available are ones built into the platforms we already use. For example, we just migrated to Alma with Primo, which has a built in researcher AI assistance. Our primary concerns with AI models revolves around accuracy, legality, bias, and accessibility. We’ve found several tools which limit results with queries deemed “sensitive subjects” which is a problem in an academic research library. While there is plenty of potential with AI and LLM tools, we’re reticent to dive in when the ethical considerations are still numerous.
-
Currently Our institution uses LMS (Blackboard) and Safe Assign software. Safe Assign is an integrated Blackboard component in it that enables to detect if students have used Gen AI for assignments. Students submit assignments on Blackboard. After submitting the AI detector, which is part of Safe Assign automatically check the entire assignment to ascertain if any Gen AI tool was used during assignment writing. If that’s the case, then Safe Assign AI detector tool will the highlight the sections where Gen AI tools were used. That information is presented to students for them to see. As the library we do also teach to be mindful and embrace these Gen AI tools but not to depend on them entirely for academic work. We also teach them about the importance of academic integrity. The Institution and library emphasis more on critical thinking been party of its mission outcomes of Higher Order Thinking.
-
Currently, my institution (as far as I know) is not using AI tools to track student engagement with scholarly content and we don’t have any current measures for evaluating GenAI for library collections.
-
How is your institution tracking student engagement with scholarly content through AI tools?
We’re not. We’re a very small staff with too much to do.What measures does your library use to evaluate GenAI models for library collections?
This can vary from person to person and is dependent on how much knowledge they possess about the technology. We have no formal methodology as a department to evaluate these tools. -
We’re still in the process of working out details re: how we’ll be using and assessing AI (we’ve got grant money for it, and our state is requiring public universities to get on board with AI, broadly speaking). At the moment, we’re not tracking student engagement, and we’re still in the early stages of sorting out our selection and evaluation practices.
-
Currently, our institution tracks student engagement with scholarly content through AI tools by surveying how students use the tools inside and outside the classroom. Though still in it’s infancy these surveys were enacted as part of a multidepartment study my colleagues and I are working on which focuses on scaffolded engagement with GenAI tools across different disciplines. We give the same explicit instruction and guide students through several modules/actives using GenAI tools per course outcomes. Students are surveyed pre/post instruction. We hope to use this data to help further AI initiatives on campus and for professional development.
In order to evaluate GenAI tools for our collection we typically attend webinars and product demos in order to get a feel for workflow and weigh it against our population, goals, learning outcomes, and benefits/concerns. Typically that involves myself and my staff testing and trialing products or turning AI features on in already existing products. Our institutional policy is still being developed and adapted so our library uses other metrics such as, cost, accuracy/relevancy, and OCUL’s AI Tools for Academic Libraries: AI Research Assistants series (blog).
-
How is your institution tracking student engagement with scholarly content through AI tools?
The institution where I work is very committed to ensuring the quality of content that can be retrieved through artificial intelligence. In fact, this is a basic job for all academic librarians. For this reason, the library designs courses that emphasize different strategies and recommendations to make the most of AI content, relating them, of course, to the career they are studying.What measures does your library use to evaluate GenAI models for library collections?
In two ways: a) through feedback from GenAI users, both professors and students; and b) through feedback from academic librarians on all of our institution’s campuses throughout the country. -
Бібліотека не відстежує активно залучення наукових співробітників до наукового контенту за допомогою інструментів штучного інтелекту. Ми використовуємо StrikePlagiarism для перевірки подібності та плагіату, а також його функцію перевірки на основі штучного інтелекту для забезпечення академічної доброчесності.
-
We still at the beginning stage of introducing AI
we have training workshops to inform students -
Discussion Questions
AI gives us a lot to think about. Share your thoughts in the Leave a Reply section below!How is your institution tracking student engagement with scholarly content through AI tools?
We still at the early stages of adopting AI. We making awareness and having training workshops collaborating with other units
What measures does your library use to evaluate GenAI models for library collections?
We not at that stage yet -
Our institution is not currently tracking engagement with scholarly content through AI tools or have any measures in place to evaluate GenAI models for library collections. The university just recently received an approved AI use policy, which the library assisted with, so now we can start to think about AI usage within the confines of the policy.
-
At our university library, we do not formally track individual student engagement with AI tools yet beyond standard database/discovery usage statistics (such as those provided in our Alma/Primo environment). Most of our current understanding comes directly from reference interactions, user training sessions, and collaboration with faculty.
When evaluating GenAI models for our library collections and resources, our main priorities are citation accuracy, source transparency, and preventing AI hallucinations. We also pay close attention to data privacy and ensuring that any AI integration serves as a complement to research rigor and critical thinking, rather than a shortcut that replaces academic integrity.
-
How is your institution tracking student engagement with scholarly content through AI tools? – Currently, we are not tracking student engagement.
What measures does your library use to evaluate GenAI models for library collections? This is something we currently are not doing.
-
Our institution is in the process of evaluating enterprise licenses so we don’t currently track AI tool usage. I don’t know if there are plans once a tool is implemented to track usage, or how they would do so. Currently, it’s by library staff through combinations of user testing, instruction, and personal curiosity.
-
As far as I know we are not currently tracking AI usage because we are just getting started with AI but our students do use Blackboard so they do track some through that platform.
We use vender platforms to evaluate GenAL models.
-
User engagement statistics are being tracked at the very least for Primo Research Assistant
-
Our library also tracks data for student uses of the ExLibris Primo Research Assistant AI (Beta) tool.
-
-
My library is not currently tracking student use or engagement with generative AI. We have a few simpler resources, such as a guide and a short video to help students engage responsibly with generative AI. The library informally researches and tests new AI features. We determine which tools would be appropriate and abide by the university’s AI and Honor Code guidelines through collaborative discussion amongst ourselves.
-
How is your institution tracking student engagement with scholarly content through AI tools? Our institution uses LMS (Blackboard) and Turnitin software. Turnitin is integrated to AI component in it that enables to detect if students have used Gen AI for assignments. Students submit assignments on Blackboard. After submitting the AI detector, which is part of Turnitin automatically check the entire assignment to ascertain if any Gen AI tool was used during assignment writing. If that’s the case, that Turnitin AI detector tool will the highlight the sections where Gen AI tools were used. That information is presented to students for them to see. As the library we do also teach to be mindful and embrace these Gen AI tools but not to depend on them entirely for academic work. We also teach them about the importance of academic integrity.
What measures does your library use to evaluate GenAI models for library collections? Accuracy, validity, structure in which information is presented, content bias, hallucination etc.
-
Where we have licenses for downloads, users cannot download AI summaries. Therefore, we position the AI services as another way of finding the most relevant content. We have started to track enquiries featuring AI produced references/citations as they are often hallucinations. An internal AI policy has recently been announced and it has been made clear that staff are not to use AI. Permission needs to be sought for projects. In terms of our content being scraped for use, the library is not supportive of collection content being used due to copyright and ethical purposes.
In my area, users seem to use AI for finding our services but come to a halt when accessing our collections. In the area That I work in, we include AI in relation to assessing market research data freely available on the internet as well as in relation to intellectual property – using AI tools for images, copy for business use. -
Tracking of students’ use of AI, we are currently using the plagiarism software that gives the percentage of AI writing.
We do not have any measures for GenAI for Library collections. Most of our collections are accessed via vendor platforms . -
As librarians for a local community college, we are incorporating the use of AI tools in teaching information literacy to students. There is a discussion on establishing a campus wide policy for the use of AI in the classroom that librarians are participating in. Most of the student and faculty engagement tracking with any AI tools is antidotal at this time. I haven’t heard yet if the full-time librarians are considering using GenAI tools in the library collections. Several of the platforms e.g. EBSCO have AI tools available for individual use. We are encouraged to use them as individuals.
-
As of now our library is not tracking student engagement with AI.
We are still developing a guide for AI use.
-
Usage Statistics for Primo Research Assistant are being kept
-
We engage personally with patrons to learn what AI they using and how they using AI.
A concern for me in evaluating an AI offering is:
1) how AI is it? Is it fluff and closer to a glorified keyword search or does it actually offer the best of AI for academia? What advantage of AI is being lost in it’s method of search?2) scope of content – I don’t really want another version of semantic scholar OA corpus. Give me something that searches subscribed content that publishers are happy with. For a start can it search my libraries subscribed FULL-TEXT content? Can it search BOOKS? Can it search CONFERENCE PROCEEDINGS? Can it search PATENTS? And if the answer is no, we are not there yet.
-
I can’t comment on the first question accurately, as my institution isn’t tracking that comprehensively.
We have rolled out Microsoft Copilot with an institutional licence for all students and staff to use, and have just added the choice of agent inside it so now users can choose to use either ChatGPT or Claude inside it. I have experimented with the accuracy of both and have found Claude significantly better in search string construction, and so will modify my teaching notes for students accordingly.
We do not enjoy the large budgets of some other institutions, so we are not adding AI assistance services to our databases at this stage, but I’m sure we would collect some data and then analyse it to see if our return on investment is there, if nothing else…added tools will be dropped if we don’t detect value for money fairly quickly. So far, the AI tools we are using inside our cataloguing are proving their worth, but there is still always a Human In The Loop with our cataloguing, and this is due to past experiences where we lightened our touch and started finding useless and outright misleading records sliding in to our collections, which wasted time and money investigating and fixing. -
My school is not formally tracking how students are engaging with AI tools. We are gathering anecdotal evidence through a particular teacher’s classroom initiative, the librarian’s (me) initiatives to test AI tools to help with research projects at the high school level. Full disclosure: we have learned a lot by tracking student use of AI tools by the academic misconduct incidents that are documented. As the librarian, I’m involved in the restorative process, and I learn a lot from the students directly. This often can be an opportunity to provide insight as to how AI tools can work effectively and expectations of use in higher education.
-
We are not actively tracking AI use. I am sure that AI tools used within PRIMO and other library databases are being tracked.
I have had a professor say he has stopped bringing his students for library instruction sessions because students were just using Chat GPT or another AI agent to “write” their papers. -
I use quite a few artificial intelligence tools, but this is the first time I’ve heard of OpenAlex.
-
Texas A&M University Libraries approach generative AI as a tool that enhances, rather than replaces, scholarly research, and they assess both student engagement and AI models through an evidence-based, librarian-led framework. Rather than monitoring individual AI use, the Libraries track engagement by collaborating with faculty, integrating AI literacy into instruction, and evaluating how students naturally incorporate AI into their academic work for tasks such as brainstorming, outlining, practice problems, summarization, and background research. These observations inform the development of workshops, online tutorials, and curriculum-embedded learning modules focused on AI fundamentals, ethics, prompt engineering, and responsible research practices while reinforcing the expectations of the Aggie Honor Code. At the same time, the Libraries evaluate generative AI models through continuous testing across multiple platforms, rigorous validation of AI-generated content against authoritative scholarly sources, assessment of methodological transparency, and careful consideration of data privacy, reproducibility, and ethical use. Particular emphasis is placed on determining whether AI tools effectively support evidence synthesis and systematic reviews without compromising research quality, recognizing that understanding how an AI produces results is just as important as the results themselves. Human expertise remains central throughout this process, with librarians critically reviewing AI outputs, developing best practices for emerging AI applications, and ensuring that generative AI serves as a complement to, rather than a substitute for, critical thinking, scholarly judgment, and rigorous academic research.
-
How is your institution tracking student engagement with scholarly content through AI tools?
As far as I know, we are not actively tracking engagement with scholarly content through AI tools in any formal way. The best way we probably do that is when people self-report said engagement when they email or otherwise interact with us.What measures does your library use to evaluate GenAI models for library collections?
We use measures set out by our institution since that guidance will always supersede ours in terms of what models are actually available to be used on our company systems. Beyond that, it becomes a budget and time & effort question in terms of how much will it take to implement and how much T&E will it take to implement and upskill on. There is also the overarching question of if our patrons and our collection will benefit from its usage. -
In our library we are using Turnitin for similarity and plagiarism and we are using its AI checking facility for academic integrity. And for many routine jobs like letter drafting, grammar checking and summary making we are using QuillBot .
-
How is your institution tracking student engagement with scholarly content through AI tools?
We are not tracking student engagment with scholarly content throug AI tools, unless this also includes the usage report generated from th eonline databases. If the latter applies, we don’t really use this data unless we need to justify resubscriptions.
What measures does your library use to evaluate GenAI models for library collections?
The insitution is currently developing and AI policy that will be used to guide how we go forward on this front. Specifically for assignments and research papers, Turnitin software, which includes AI detection, is used. -
As professionals in LIS, we are trying to have the place that corresponds to us in the institution and in the treatment of important issues (such as the use of AI) that demand a framework and regulations.
-
As far as I’m aware, we’re not tracking student engagement.
We’ve developed a mechanism for assessing new AI tools in library resources specifically, with heavy reference to the hard work already performed on this front at Rutgers. But with the current state of things it feels a bit moot – most AI tools we have encountered so far either can’t be disabled, or are on by default and can only be disabled by individual patrons (in some cases they have to specifically create an account with a third party to do so).
Even assessing the likes of Perplexity and Gemini (currently considered outside of our scope) and presenting our findings to students in their one or two library/research classes would be of limited use when teaching staff outside the library are telling hundreds of students at a time that “ChatGPT is great, I use it for all my research/coding/project management and so should you!”
-
My institution ist not tracking student engagement with scholarly content through AI tools.
The university has an AI framework and the library offers courses on AI tools for literature research. Regarding genAI it is a topic of another service institution on campus which offers information and courses on scientific writing.
-
Our library is not currently tracking student engagement with regards to AI tools.
We are still in the process of formulating our AI policy for the library. and how we would like to ensure that there is a high level of accuracy and accountability when using AI for research purposes. -
We are not tracking student engagement with any AI tools or how they use them for research.
The following core values – Accountability, Access, Autonomy, Assurance – guide our evaluation and integration of GenAI tools when selecting new products or considering whether to enable them within existing licensed academic databases/collections.
Leave a Reply