Generative AI in Libraries (GAIL) Virtual Conference

This is a tentative schedule, subject to change.

Day 2: June 12 Agenda

Please note, all times are listed in Eastern Time on this agenda.

1–1:45 p.m.

Lightning Round 1: 15-Minute Presentations

1 - 1:15 p.m.

In most history courses, students are expected to engage with primary sources, defined by the RBMS-SAA Guidelines for Primary Source Literacy as “materials in a variety of formats that serve as original evidence documenting a time period, an event, a work, people, or ideas.” Over the course of a history degree, students develop skills to critically evaluate a wide range of primary sources, by, e.g., considering the purpose and audience of a given source, as well as the perspective of the creator(s) and their possible biases.  

Emerging technologies such as generative AI have the potential to disrupt the development of primary source literacy by creating fake primary sources that mimic the tone and language of historical actors. However, “fake” sources have always been part of the documentary record, ranging from historical fiction to tall tales or spurious stories – just think of the story of George Washington and the cherry tree and other historical myths.  

To incorporate AI literacy into primary source literacy, a librarian developed an innovative one-shot workshop for an upper-level history course that combined analysis of “true” primary sources, “fake” (i.e., AI generated) sources, and spurious stories drawn from 18th- and 19th-century British North America. Students were encouraged to consider the authority, tone, purpose, and audience of materials, as well as their knowledge of the time period, to identify which were real, fake, or spurious. Integrating generative AI technology into traditional classroom instruction of evaluating sources allowed students to develop AI literacy alongside primary source literacy, without requiring an entire class session devoted to these new technologies. This activity can also be adopted into other contexts, such as by focusing on academic or secondary sources rather than primary sources, or could be used as the basis for collaboration between librarians and interested professors to promote AI literacy.

1:15 - 1:30 p.m.

Though there is much conversation about the impact of generative AI on higher education, there is little information on how college students are actually using AI. There are a range of helpful generative AI uses regarding commonly-fielded questions in academic libraries, such as the research process. These uses include brainstorming research topics, identifying search terms, developing outlines, summarizing text, and proofreading, among others. However, generative AI also poses a threat to academic integrity. For example, ChatGPT has been cited as a disruptor to the undergraduate essay, it cannot be detected by most plagiarism checkers, it can respond to tailored prompts such as "write in the style of a college student," and it can be biased, inaccurate, and hallucinate fictional citations.

While the positives and pitfalls of generative AI use at the college level are evident in the literature, students’ knowledge of this information remains unclear. The extent to which students are using AI for personal and academic uses is also largely unknown. In an effort to better understand the realities of students’ AI usage, an anonymous survey was distributed to undergraduate students across all disciplines at one institution during the Spring 2024 semester. This presentation will discuss the findings of this survey, including quantitative and qualitative data on students’

1. overall usage of generative AI,
2. usage of AI for college coursework, and
3. opinions on ethical uses of AI. 

This information will help librarians gain insight into:

1. students’ understanding of ethical usage of AI,
2. students’ usage of AI throughout the research process, and
3. students’ awareness of potential negative outcomes in using AI.

Survey responses will be explored in connection with implications for reference and instruction in libraries. Comments which express using AI for student support, such as assisting with English language skills, will be highlighted.

1:30 - 1:45 p.m.

From a research paper to a systematic review, librarians develop search strings for a variety of databases depending on the topic and the field of interest.  Generative AI has been used to develop computer code as well as to tweak code to have it run more efficiently.  Just as generative AI has been used for computer code, it can also be used to develop a search string.  The presenter will review different strategies for writing prompts to write a search string for a research paper as well as for a systematic review.

1 - 1:45 p.m. 

The launch of GPT-4 in March of 2023 led to a lot of buzz about artificial intelligence, much of it fearmongering and misinformation. With rapidly evolving, competing reports on the impact of AI (that it will replace - or assist? - workers, that it will exacerbate - or solve? - the climate crisis), patrons and librarians alike expressed concerns about exploring generative AI tools on their own. Due to the lack of trustworthy resources to help users navigate this rapidly evolving technology, generative AI emerged as an urgent site of both information literacy and skill development. 

Libraries and librarians are uniquely positioned to stem the flow of misinformation, alleviate the culture of fear, and to provide classes and tools that aid in skill development around new technologies. As passionate advocates for digital equity, we created a class that would render AI approachable and accessible. This biannual class addresses the basics of artificial intelligence, useful applications for life and work, its limitations and ethical implications, and discussions of the possible futures of AI technology. Through the removal of real and perceived barriers, we help users navigate the minefield of information about Generative AI, so patrons are empowered to draw their own conclusions and to use AI as they choose. 

This presentation will cover the process of researching, planning, instructing, and updating this biannual class. We will discuss how we used several Generative AI programs to familiarize ourselves with their operation. We will demonstrate AI's value as an instructional and research tool, detailing how Generative AI was used to help develop our lesson plan. We will share how we refresh our research to ensure we are dispelling misinformation, sharing accurate and up-to-date information, and demystifying artificial intelligence so that it is accessible to all.

The word salon is French, originally meaning "reception room." In 1800's France, the meaning grew to include a "gathering of elegant people" occurring regularly in such a room. Rather than focusing on building skills or producing deliverables, it encourages philosophical debate and engagement in current events and ideas as they expand.

In 2023, George Mason University Libraries initiated an internal summer salon series, inspired by historical science salons as forums for intellectual exchange. This series aimed to cultivate open dialogue among library staff regarding the impact of artificial intelligence (AI) tools in higher education and society. The salons encouraged philosophical discourse and engagement with contemporary issues, particularly surrounding technologies like ChatGPT.

Over three months, six scholarly discussions delved into the effects of AI tools on libraries, higher education, and society. Subject librarians facilitated these community-building sessions, both online and in-person. The success of the summer series led to its expansion into a campus-wide initiative for 2023-2024, covering diverse topics such as "AI in the Social Sciences and STEM," "AI in Business and Government," "AI, Ethics, and Academic Integrity," and "AI, Biases, and Prejudices," among others. Recognizing the need for further opportunities for discourse, the organizers transitioned to offering the salons in both in-person and Zoom-only formats. This shift toward a salon-style approach proved pivotal in fostering comprehensive conversations and involving the broader university community in exploring AI's interdisciplinary and intersectional impacts.

This presentation will focus on adopting a science salon format for discussions on ChatGPT and similar tools. It will outline the outcomes of the salon series and its potential applications in libraries of all sizes, addressing marketing strategies, assessment methods, and providing recommendations based on insights gained from these gatherings. Real-time accessible slides will be available for participants to download during the presentation.

AI technology is a tool that will become commonplace in the near future. It is our responsibility to learn how to use this tool, and to use it wisely. My proposal is for a 45-minute seminar that covers the potential dangers and weaknesses of AI. It will be broken into two parts. 

The first part will deal with workplace safety and technological responsibility. It will cover topics such as data security, what information can safely be shared with AI systems, and the possible legal ramifications of misusing AI, even accidentally. AI is a product, and companies are creating excitement because they want you to use it, and they want you to buy it. Furthermore, if you are not the customer, then you are the product. Many companies are using AI to gather data, which can then be sold on the internet. We must therefore be careful to be responsible stewards of our patrons and organizations private information.

The second part will deal with social responsibility. AI is not an unbiased source of information. It can only be as reliable, and as unbiased, as the data that is fed into it. AI can be manipulated and controlled by individuals and organizations with private agenda’s. Furthermore, it can be influenced by cultural bias. AI is not a peer reviewed and reliable source of information. It can draw conclusions that are culturally biased, and lead to unequal outcomes for marginalized people. It can, and has, propagated dangerous and hurtful stereotypes about marginalized groups. Furthermore, AI has been known to make up fake data sets in order to fulfill requests. These digital “hallucinations” can skew data and misinform the public. All of these are dangers that we, as educators, must be aware of.

Everyone’s talking about AI! It’s on the news, featured in webinars and conferences, and popping up at the reference desk. But for many in the library profession it still has not actually hit our desktops or workflows. In this session, we will provide guidance and tools to start developing your own AI literacy by identifying common terminology and useful resources, providing a brief overview of the current landscape, explaining how tools like ChatGPT work (and why they are so popular), and offering a framework to support your own learning. Attendees will build their confidence and conversational fluency around AI, enhancing their ability to evaluate AI products and database integrations and support library users as they explore and leverage AI tools for their own information needs. AI is for everyone, make it work for you!

2–2:45 p.m.

Lightning Round 2: 15-Minute Presentations

2 - 2:15 p.m.

Finding topical historical primary sources, which exist in a range of proprietary library databases and free, online digital libraries, can be a challenge for the most intrepid History librarian, nevermind students. Can AI make this easier? Some initial testing quickly revealed the limits of closed AIs, but potential with Perplexity.ai, an alternative search engine and large language model that summarizes search results at a scale much faster than an individual librarian could do. And, it cites its sources. With a proper prompt, Perplexity seems adept at doing a large scale internet search and then summarizing the search results, with links back to its sources, which include library research guides and government information resources. This presentation presents preliminary research on the utility and limits of Perplexity as a tool to find information on and links to historical primary sources on a range of topics that a typical undergraduate student or History major might explore, along with suggestions for how we might use this tool in library reference and instruction.

2:15 - 2:30 p.m.

Creating content for social media can present a significant challenge for librarians tasked with managing their library’s online presence due to time constraints, lack of inspiration, or difficulty condensing information about library resources. If you’ve developed a brand kit, including elements such as themed days of the week as well as brand voice, colors, and fonts, leveraging generative AI tools like Canva’s Magic Studio and ChatGPT can become a staple in your content creation. These tools can play a crucial role in generating post ideas—for instance, you can ask ChatGPT for historical events or notable figures born on a specific day—or in summarizing and rewriting content to ensure consistent brand messaging in posts. This presentation will cover the various ways AI can become a go-to resource for library content creators.

2:30 - 2:45 p.m.

Consensus is a search engine that uses AI to make academic research more efficient and effective. The goal of the presentation will be to introduce our product, our unique approach, discuss generally how AI can be best used to augment research and conclude with a demo of our product.

2 - 2:45 p.m.

In a world where the nuance of sustainability has become critically important, the intersection of Library instruction and AI utilization has made way for adventurous information seeking, analysis and use. Integrating sustainability across the curriculum involves getting faculty excited about meeting students where they are (on ChatGPT) while fostering information and AI literacy. Both of which may be topics that they themselves use, but don’t fit within their subject matter expertise. Librarians can provide instruction that not only benefits the students, but gives faculty a connection point that resonates with course level learning outcomes, deeply. 

Semantic Search (aka Vector Databases) have driven a massive increase in the widespread availability of effective search tools. However, the non-deterministic nature of these tools can make them unsuited for common academic tasks like systematic literature reviews. How can researchers take advantage of better search, while still having reproducible results? We'll explore tools like query expansion (SPLADE) and AI assisted screening, which can be powerful tools towards this end. 

Accessibility, Equity And AI: What's New and Where Are We Going?
Robert Gibson, Dean, Instructional Technology & Academic Services, Wichita State University Campus of Applied Sciences & Technology (WSU Tech)
Ethical Considerations & Policy in AI Track

This presentation explores the evolving landscape of artificial intelligence (AI) in the context of educational accessibility and equity. It delves into the latest advancements in AI that aim to level the playing field for students of all backgrounds and abilities, while also discussing the ethical considerations and challenges that arise. The presentation aims to offer a balanced view of how AI can be harnessed for greater inclusivity in education, while also scrutinizing its limitations and potential for exacerbating inequalities.

Generative artificial intelligence (Gen AI) tools such as ChatGPT have proliferated throughout the educational landscape faster than many educators can adapt. These emerging trends have made it essential for library users to hone their information literacy skills.  In response, the presenters have developed a set of dynamic lessons focused on the intersection between GenAI and information literacy as a means to develop user skills.  This session was designed in a ‘plug and play’ format to account for multiple instructional settings, relevant frameworks, and pedagogical methods.  The resulting lesson includes three active learning activities, tied to outcomes from ACRL’s information literacy framework.  The module was shared via a staff intranet to allow other library staff members to utilize and adapt these activities without needing extensive GenAI expertise.  This presentation will discuss the development, philosophy, and implementation of the lesson module, with emphasis on encouraging the wise and productive use of GenAI.

3 - 3:45 p.m.

Lightning Round 3: 15-Minute Presentations

3 - 3:15 p.m.

This session presents the results of a year-long study applying AI tools to the William Elliot Griffis collection at Rutgers University Libraries. The study builds a use case for the application of AI to manuscript collections to improve access and enhance discovery, allowing comparisons across disparate materials. It provides an in-depth analysis of the authors’ experiences with AI tools, including generative AI software, that provide a useful guide to others looking to explore AI for their collections. Twenty-six tools designed for developers and end users, librarians and scholars, were applied to the untranscribed handwritten and typewritten documents and photographs in the William Elliot Griffis collection at Rutgers University Libraries.

The tools produced narrative descriptions, identified hidden patterns, and suggested new ways to organize the material. Session attendees will come away with knowledge of what types of AI tools are available that are relevant to manuscript collections, some of the processes involved in their use, and their comparative strengths and weaknesses. The purpose of the session is to provide an understanding of practical ways that AI can improve access and enhance discovery of distinctive collections. Barriers to the use of AI include the need for Infrastructure support and programming expertise.

3:15 - 3:30 p.m.

"Let’s see what ChatGPT has to say about…" This may be a common refrain at your institution. It definitely is at ours! For some, including students, generative AI has become a staple in their everyday lives; whereas, others may currently view it with a more skeptical lens. One thing we all know is that there’s been a proliferation of generative AI tools from which students and researchers can choose. Business librarians at a private university thought it crucial to teach students effective strategies for evaluating information using both generative AI and scholarly library resources. The verdict: students walk away with a better understanding of the benefits and challenges of incorporating generative AI as part of their research toolbox. Hear how one librarian incorporated ChatGPT and the SIFT method into business information literacy instruction!

3:30 - 3:45 p.m.

Writing or updating procedure manuals is a daunting task, especially when it involves countless screenshots and meticulous documentation. But what if you could streamline the entire process with the click of a button?

Enter Scribe.ai: the revolutionary tool that automates the creation of step-by-step guides for any computer application. Imagine no more tedious screenshot capturing or video pausing. With Scribe.ai, you get instant, precise screenshots coupled with concise action descriptions.

Why settle for less when you can enhance the accessibility of computer instructions? Scribe.ai not only simplifies the learning experience with visual handouts but also caters to diverse learning styles with tailor-made directions.

3 - 3:45 p.m.

The prevailing sentiment among Humanities faculty on our campus is that Artificial Intelligence (AI) merely serves as a "plagiarism machine." This perception echoes the skepticism that greeted the internet in the 1990s and Wikipedia in the 2000s. However, just as with previous technological advancements, integration and critical engagement can lead to progress and enrichment. Banning AI doesn’t help our students think about their careers after college, or their life-long learning. Now is the time for exploration and experimentation, particularly in fields often considered traditional and conservative in light of new technologies. 

This presentation illustrates how generative AI can be used ethically and responsibly in the Humanities as a:

1. Research assistant
2. Summarizer
3. Data extractor
4. Text transformer
5. Tech support specialist
6. Visual Illustrator
7. Writing Assistant

This is information I share with teaching faculty as part of our discussions around information literacy instruction classes with the goal to demonstrate AI’s potential role in refining analysis, fostering creativity, and facilitating discovery. In two early success stories, I will share how an English lecturer used A.I. as a summarizer in her first-year composition class, and how a History professor employed AI in an upper-level class to analyze primary sources. 

This presentation will be useful to librarians across library sizes and types and will be made engaging and accessible using relatable examples, accessible presentation slides, clear visuals, and an accompanying handout (if possible).

By applying generative AI with careful analysis and ethical guidelines, faculty in Humanities fields will undoubtedly discover new research and scholarship opportunities. Early adopters will help to foster a harmonious relationship between this innovative technology and their enduring quest for knowledge in the humanities.

Arizona State University (ASU) recently partnered with OpenAI to launch a Spring 2024 AI Innovation Challenge, where groups could propose various experiments to explore how generative AI is shaping the future of learning, research, and work in higher education. Those that were approved gained access to ChatGPT-4 licenses for approximately 8 weeks, experimenting with the system to encourage growth in knowledge, learning, and academic excellence. As AI becomes ubiquitous in higher education, librarians, who have unique takes on issues like authority, relevance, and accuracy, should have central roles in assessing these tools. ASU Library is participating in the AI Innovation Challenge to advance AI literacy skills and knowledge about AI’s uses, while also acknowledging its limitations across all libraries.

ASU accepted two ASU Library proposals, creating two teams to complete the projects. An evaluation framework project investigates whether we can develop a source evaluation framework (like CRAAP or SIFT) to assess text generated by ChatGPT. A personas project involves creating AI personas to simulate different patrons with various research skill levels to train new personnel for ASU Library’s “Ask a Librarian” chat service. In both projects, which are expected to wrap up in May, the teams are learning how best to use ChatGPT. 

This panel will reflect on the two projects’ goals and outcomes, including whether we can create an effective framework and whether AI can train new library employees. Discussion will offer insights into obstacles we encountered and critical issues that must be acknowledged when working with ChatGPT, like bias, diversity, hallucinations, “truthiness,” and data sourcing. Panelists will share their various experiences with ChatGPT and how they approached these issues and others. Furthermore, this panel will focus on AI’s influence through lenses of reference behavior, instruction, representation, and knowledge sharing. Lastly, if time, attendees can pose questions to panelists.

In early spring, the presenters were invited to host a workshop for a group of 30 PhD students taking part a “Dissertation Bootcamp” organized by the university’s Graduate School. This presentation will describe the support we offered the PhD students during the workshop, as well as incorporate some of the questions the students asked during the session. We will highlight the guidance we provided related to relevant issues and concerns when using as a PhD student, including copyright/intellectual property, adhering to IRB protocols, reproducibility, bias, preparing for the defense, and establishing one’s own ethical boundaries.

In addition to offering guidance on how to choose appropriate AI tools (and what to look out for), we will demonstrate several generative AI tools that could be most helpful for students in the dissertation-writing stage, including Perplexity AI, Keenious, Scite.AI, Research Rabbit, YouPro for Education, and Elicit. We also advice PhD students on how to enhance their research using AI-powered library resources. These tools can help them discover relevant materials, analyze data more efficiently, and gain deeper insights from their studies. Also, we will briefly discuss tools being developed by publishers such as Clarivate and Elsevier. We hope to use remaining time for questions and/or discussion.  

This presentation will appeal to library workers who are interested in incorporating Artificial Intelligence (AI) generated images into their work.  Much has been written about the limitations of images produced using AI. We will talk about how we successfully – or mostly successfully-- incorporated AI images into a professional presentation that covers a sensitive topic.  That presentation (forthcoming) focused on how larger-body individuals navigate situations and spaces in libraries and on university campuses.  We decided to incorporate AI images in the presentation because it was less intrusive than using actual photographs of large-bodied people.  In today’s presentation, we will walk participants through our approaches to generating these images using DALL-E.  Many of the generated images were unacceptable to us because they were disrespectful (cartoonish and rooted in stereotypes), but with persistence, we were able to generate usable images for most of the situations we wanted to depict.  In this session, we will talk about prompts that seemed to work best and those that didn’t work well.  We will discuss AI ethics, fair use, and biases. Our presentation will be useful for librarians who are interested in incorporating AI images into instruction materials, library publications, library web pages, and outreach activities.

This is a tentative schedule, subject to change.

Day 3: June 13 Agenda

Please note, all times are listed in Eastern Time on this agenda.

1–1:45 p.m.

Lightning Round 1: 15-Minute Presentations

1 - 1:15 p.m.

LibGuides are indispensable tools for librarians, facilitating information literacy teaching in a variety of contexts and enhancing the overall learning experience for users. This is particularly true for guides tailored to specific courses. Course guides and others that closely align with student learning objectives and assignment requirements, provide a more immediate and targeted response to user’s needs. Yet, developing guides with such customized content often proves to be a time-consuming process for Librarian creators. 

Librarians might consider leveraging generative AI to assist them in creating guides with more focused and timely content, making their work easier and more efficient. Generative AI tools can assist in implementing best practices, from formulating searchable, timely topics to refining language for clearer, user-friendly content. AI tools can also be helpful in suggesting visual elements that complement course themes and cater to diverse learning styles. While AI continues to evolve, with its full implications yet to be understood, it holds the potential to simplify a facet of instructional duties. Librarians are thus able to devote more of their energies to high-priority responsibilities.

This presentation will demonstrate how a teaching librarian has employed AI to guide the creation of customized course guides. Current AI applications will be discussed as well as those under consideration for future implementation, with a specific focus on enhancing student accessibility.

1:15 - 1:30 p.m.

During the Spring 2023 semester, librarians from Lehman College’s Leonard Lief Library began a research study, which analyzed transcripts of chats received by our Ivy AI chatbot. Powered by OpenAI’s ChatGPT API, the chatbot has been a staple on the library’s homepage since 2019. The goal of the research study was to better understand the types of questions the bot received and use this data to improve the User Experience (UX) of the bot and library web services for the college’s students, faculty, and staff.

One challenge that emerged was that patrons were struggling to find course materials from the library’s course reserve collection when using the library’s discovery system, Primo VE. Questions about course materials are among the most popular reference questions the library receives, both at the reference desk and on library chat. When patrons asked the chatbot whether a specific book was available or not, the bot could not consistently provide this information. This finding, along with an expressed desire from librarians for better tools to help patrons find textbook information for students motivated our library to find a solution. 

Responding to this need, the library created a dedicated webpage with a list of reserve books to be indexed by the chatbot, designed to improve the bot’s discovery ability and assist users. The authors worked with the vendor Ivy to format and ingest the list to ensure maximum exposure and accuracy.

This presentation will demonstrate the response patterns that led to the creation of the reserve list page, and how the book list improved the discovery of library materials. It will also discuss what challenges were encountered, and how other libraries can use their own chatbot data to respond to student UX pain points and use AI functionality to implement changes and increase usability. 

1:30 - 1:45 p.m.

This presentation aims to explore the influence of input language on generative AI outputs and its implications for information literacy. We will investigate how biased versus neutral search language may affect the outcomes generated by AI models. By examining this phenomenon, we can gain valuable insights into the effect of using different search language in various information gathering scenarios. The presentation will offer examples to illustrate how biased language may affect information accuracy, credibility, and user perception. Additionally, we will discuss strategies to enhance information literacy through the use of neutral language.

1 - 1:45 p.m.

This session aims to offer insights into the transformative power of AI in education, from teacher support to library innovation and mental health advocacy. This presentation will explore how AI tools are revolutionizing teaching practices and fostering innovation in school libraries. Magic School AI, KhanMigo, and Canva platforms are examples that can support teaching practice and will be discussed. We also discuss how Retrieval-Augmented Generation (RAG) provides ways to limit AI hallucinations and incorrect returns on queries. ISTE’s StretchAI will be used as an example.

Gen AI can support teachers, from lesson planning and assessment to fostering meaningful engagement with students. We will discuss how AI can support new teachers and nurture the librarian-novice teacher relationship, thereby instilling resilience and limiting burn-out. We will also highlight innovative uses of AI in school libraries, emphasizing evaluation methods and AI literacy lessons that have successfully been implemented. Gen AI has a role in promoting mental health awareness as well as the opportunity for timely interventions. We expand on the notion of mental health support using AI. Through interactive demonstrations and real-life examples, our presentation aims to inspire educators to embrace AI as a valuable ally in their teaching journey. 

We discuss how to implement guidance and policy, imperative as guardrails as well as supports for the use of Gen AI. We will use specific examples to help frame this important aspect of GenAI use.

At Minnesota State University - Mankato (MNSU), we’ve undertaken several experiments and initiatives focused on Generative Artificial Intelligence. At the start of the fall semester, we collaborated with university Information Technology Services to present a professional development session for returning faculty through the MNSU Center for Excellence in Teaching & Learning on “5 Tips for Teaching with AI.”

We also presented to librarians across the regional consortium, Minitex, on “The Library & Generative AI.” This presentation included several demonstrations. It was offered as an introduction to Generative AI focused on topics most relevant to librarians, including information literacy, as well as copyright and license-related concerns. Later in the fall, we offered a daylong “Experience Friday” workshop for area high school students on “AI at the MNSU Library.” The workshop included sections on basic prompting, responsible use, learning with AI, the future of work, and the dark side of AI.

Throughout the fall, we experimented with using chatbots to help with licensing as described in the NASIG Fall conference presentation, “AI as a License Review Assistant.” Finally, leading into the spring semester, we also experimented with using Dall-E to develop collections data illustrations as described in the SUNYLA conference presentation, “Novelty Visualizations of Collections Data: Real Impact or Comic Interlude?” At the GAIL virtual conference, we will review these initiatives and others to consider what has worked best and what we might not repeat. As we look to the future, we are asking ourselves, where should we focus our efforts?

Amid discussions of AI and machine learning tools and their impact on contexts of university teaching and library services, a common narrative has emerged: students are using AI tools extensively, a majority have acquired substantial expertise in AI tools, and the rest of the academic community is falling behind and must strive to catch up. 

But to what extent do these premises stand up to scrutiny? We sought the perspectives of students themselves via a survey whose findings challenge these prevailing assumptions. This presentation will discuss the survey results and their implications, highlighting the need to embrace a more nuanced approach to AI adoption in instruction and in the provision of library services. We conclude that we need to carefully assess assumptions about students’ AI proficiency and leverage librarians’ unique position to solicit student feedback about use of AI tools; that the inclusion of student perspectives in shaping AI adoption is vital; and that students are not uncritical adopters of AI and have nuanced insights and critiques that are evident in their choices about when and how to deploy AI tools to assist learning.

Discover how strategic prompt design can help you harness the power of generative artificial intelligence (AI) in your library.  Through a series of examples, the presenters will demonstrate the impact that well-crafted prompts can have on the quality and relevance of AI-generated outputs.  Attendees will learn how to create prompts for successful AI outputs while mitigating common AI issues of bias and hallucinations.  Real-world examples will include how to curate legitimate reading lists, craft MARC records, enrich metadata, and generate inclusive images.  Unlock your library’s full AI potential to enhance user experiences and patron engagement.

2–2:45 p.m.

Lightning Round 2: 15-Minute Presentations

2 - 2:15 p.m.

Generative AI has become ubiquitous within library and education conversations since the release of OpenAI’s ChatGPT about 16 months ago. Immediate and understandable fears arose regarding academic integrity and misinformation, as ChatGPT seemed to gain unprecedented popularity virtually overnight. Just as with any emergent technology or tool, it is the role of librarians to determine methods to incorporate information literacy instruction into the ChatGPT conversation, ensuring that users are well equipped to use generative AI in appropriate contexts. So, what is an appropriate context for using ChatGPT, or other generative AI? I have experimented using AI for assignments in my one-credit library course (LIB 102) at the CUNY College of Staten Island, a diverse public college in New York City.

In this presentation, I will demonstrate how I have used ChatGPT to assist me in generating activities that help students understand proper and improper use of generative AI in their research. I compare Google’s Gemini, Microsoft Copilot, and ChatGPT to track hallucinations with citations and information sources, and use this as a teachable tool for students learning the differences between information sources and chat bots which synthesize and conversationalize data that they’ve been fed. I will share how ChatGPT can help students develop research questions, choose search keywords, and help pinpoint a specific topic to research, and recount how my LIB 102 students practiced this. I’ll also share the prompts I’ve used in ChatGPT to help use it as a tool to brainstorm in-class activities to prepare students to use generative AI as a support tool, while recognizing its limits, scope, and intended use. Questions of bias, plagiarism, and ethical information use will also be addressed.

The advent of new technology has always heralded the perception that the demise of librarianship is imminent. Recent significant waves of technological change that come to mind include the transition to online “card” catalogs, the implementation of the World Wide Web, databases delivering online content, and now generative AI. With each technological change, librarians have adapted and thus moved the profession forward keeping libraries at the heart of their communities. Generative AI has the potential to enrich our professional frameworks, specifically around Information Literacy and Scholarship as Conversation. It is offering new ways for stakeholders across campuses to interact with information, expand their perspectives, and contribute their voices to the ongoing scholarly discourse. Once again librarians should be at the center of this conversation as we help to structure how we focus our attention on generative AI.

2:15 - 2:30 p.m.

This presentation is focused on sharing the results of an ongoing research project led by the presenters.  This project is designed to explore the impact of generative AI tools, specifically ChatGPT, on both the information literacy and the information-seeking behavior of engineering students.  This research is being conducted through a “think out loud” exercise where students engage directly with ChatGPT as well as the open internet to answer a multipart, complex engineering question.  The presentation will focus on the iterative process used by the researchers to develop the complex engineering problem at the core of the “think out loud” exercise, as well as share some of the preliminary themes discovered in the data from the first set of participants.

The research team spent much of the 2023 fall semester developing a complex engineering question that would allow the team to gather data on how participants leverage ChatGPT as it relates to a number of concepts in information literacy and information-seeking behavior including discovery, authority, and synthesis.  The research team used a backward design approach to craft a question that is directly related to the design thinking process used within engineering. 

Using a “think out loud” observational session coupled with audio and screen recording, the researchers have started collecting data directly from undergraduate and graduate students in various engineering disciplines engaging with this complex question. The researchers will present the preliminary data from the first completed observations.  Using grounded theory to analyze this data, the researchers will also share some of the themes seen in the data.

When a new technology seems set to disrupt the information landscape, affecting libraries no less than the field of education broadly, it is understandable to feel as though the disruption puts everything we know into question. The fact is, libraries have weathered disruption after disruption in the last 40 years, from computers to the internet, social media and the ability of the citizen scientist to gain prominence over scholars backed by tenure and peer review. We have tools with which to approach the disruption of generative AI and its impact on library services and higher education. One such tool is the theoretical framework presented by Ranganathan’s Five Laws of Library Science (1931). With small shifts in language, we can update this framework to apply to modern information challenges in a way that preserves the core values that ground our practice. Guiding our users to the right resource for them, whether they are incoming freshman or tenured faculty, is extremely dependent on the context in which the resource is used and the end product that is envisioned. In order to uphold the rigor of academic publishing, librarians need to thoroughly understand emergent technologies and ground them into our foundational theory to prevent the academy from perishing.

2:30 - 2:45 p.m. 

Library research guides are crucial in helping students develop research and information literacy skills. This presentation reports on a study conducted between November 2023 and February 2024 that focused on AI-related library research guides at R1 institutions. The study was guided by the principles of AI Literacy and the ACRL Framework for Information Literacy, and its objective was to explore the practical and ethical dimensions of AI Literacy as articulated in library research guides. The methodology of the study involved content review and data extraction. 

In this presentation, we explore the presence or absence of instructions on creating prompts to elicit desired responses from language models such as ChatGPT. It also examines how these guides engage users with topics like human biases (such as racial bias) and machine-based biases (like algorithmic transparency). This presentation highlights these empirical findings and lays the groundwork for best practices in crafting library guides that actively and critically engage with Generative AI tools that promote information literacy principles.

Library guides have the potential to empower students to better understand the broader societal implications of AI and develop a more comprehensive understanding of its multifaceted role in today's academic landscape. These principles will extend beyond the immediate scope of AI-related guides, advocating for a holistic and inclusive approach to information literacy in the age of rapidly evolving technologies.

This presentation will explore the intersection between AI literacy, intellectual property (IP), and libraries. The session will discuss AI application in libraries and the information organization. Topics discussed will include the user experience, user interface, and neural networks and how they comprise the backbone of AI application user engagement front to backend. In doing so, the speaker will explain the implications between the user, the constraints of IP and its application / implementation as they apply to libraries and resource-rich data depositories. The session will conclude with a short discussion of (the above-mentioned) as it relates to diversity with an emphasis on people with disabilities (close captions provided), including the neurodivergent.

2 - 2:45 p.m.

This presentation reframes the library one-shot as an opportunity to engage students, Faculty and community patrons by casting them in the role as co-creators. Using a multitude of Ai based web applications and games, attendees will get a broad overview of how to leverage these programs to teach critical thinking and literacy skills. A major challenge to chunking up Ai is its speed of adoption and ubiquity in contrast to its opaque construction. This becomes even more of a challenge when considering the digital divide and the intersectionality of information borders.

The goal of this modular presentation is to use Ai as a vehicle for discussing information search skills; data safety practices; the benefits, harms and risks of large language API’s, accessibility gains and deficit models; contextualizing Ai and other writing tools as extensions not replacements; centering writing as critical thinking and the importance of TASL attribution in light of the tenuous copyright status of computer assisted output. No previous coding, GAN or huggingface experience required!

My presentation will examine academic opinions regarding Generative AI, its potential uses, and potential ethical concerns regarding the tool. Attendees will also be able to use Generative AI tools in a constructive and effective manner.

In order to demonstrate the abilities of generative AI, attendees will participate via laptop in a number of activities (and discuss them in small and large groups), including: creating increasingly precise inquiries, using ChatGPT to create assessment tools using Canva AI image creator to create images for programming and/or personal use, examining Claude responses to learn about the AI’s syntax and habits, using ChatGPT to formulate responses to assignment prompts, and fact-checking Bing AI.

Most of the time will be involved with discussions, experiential training, and/or demonstrations. Attendees will be able to use generative AI tools or see them demonstrated in an effort to learn how to use them in a constructive manner. This training will be focused on how to use generative AI effectively in a wide variety of settings, demonstrating value to all types of peoplel

Through these experiences, librarians and staff will learn both about the subjects that their patrons want to research and how to effectively use technology for non-research purposes. They can apply the lessons learned through generative AI-enhanced programming as they augment their work products and trainings through the use of text generators, image diffusers, audio generators, and other tools. They can provide complex prompts and learn how the chatbot will react to certain phrases.

• Takeaway 1: Evaluate the potential of generative AI in creating library resources through personal experiences.​
• Takeaway 2: Understand the importance of librarians' attitudes towards technology in shaping students' ethical use of these tools.​
• Takeaway 3: Understand and apply the TCoP model of technology integration in the context of generative AI.​

This workshop will introduce attendees to the use of Large Language Models for Colloquial Data Science. Colloquial Data Science is the use of natural language prompts to solve data science problems. While many in libraries have at least a passing familiarity with basic statistics, such as measures of central tendency (mean, median, and mode) and variation (standard deviation), fewer librarians or library staff are comfortable with programming. Large Language Models such as Anthrophic’s Claude, OpenAI’s ChatGPT, and Google’s Gemini excel at code generation. The workshop will cover the basics of using an LLM for performing exploratory data analysis—including generating summary statistics and visualizing various variables—of a dataset.

3–3:45 p.m.

Lightning Round 3: 15-Minute Presentations

3 - 3:15 p.m.

Science and engineering librarians from MIT Libraries will report on a workshop we designed and delivered about evaluating generative AI tools for scholarly research. The workshop showcased AI tools for finding and evaluating scholarly literature, summarizing and explaining documents, and assisting with subject-specific research tasks. We will describe how we designed our framework for evaluating generative AI tools from user perspectives. We will also share examples we used during the workshop to illustrate the application of our framework.

We designed the six elements of our framework to guide learners how to evaluate AI tools based on their purpose, training data sources and algorithms, costs or other access barriers, result accuracy and comprehensiveness, time investments, and impact on learning.  

This presentation will also explore several elements of library instruction about generative AI tools. These include balancing how much technical background information librarians should know and be capable of explaining, devising approaches for evaluating generative AI tools that are appropriate for students and researchers from user perspectives, and implementing collaborative learning about generative AI. We will also share our reflections on the workshop’s contribution to AI literacy instruction and future directions we might take.

Learn how one academic library is taking a transformational approach to exploring the possibilities with artificial intelligence (AI) through the creation of an AI Interest Group. With AI rapidly reshaping the library profession and higher education, our AI Interest Group aims to equip all library employees with the knowledge, skills, confidence, and mindset to embrace AI responsibly and consider leveraging its potential in their library work. 

Since October 2023, the AI Interest Group has served as an inclusive forum that has brought together colleagues from roles across all library departments and with varying levels of familiarity with AI. Topics have included generative AI, social and emotional intelligence, bias and fairness, and the considerations involved in the ethical governance, policymaking, and regulation of AI. We even delved into discussions about the ownership, authorship, and attribution of AI-generated content. 

We’ll share how we gathered and prioritized ideas and subsequently offered programming to learn from one another through thought-provoking discussions, webinar watch-parties, interactive demos of specific AI tools, guest speakers from different disciplines on campus, and sharing experiences of using AI in our work.  

We’ll also highlight key considerations, including challenges, successful outcomes, and insights about launching and sustaining an engaging AI Interest Group that maintains high engagement each session. Attendees will leave with ideas, considerations, and a roadmap for establishing a similar grassroots initiative. 

Join us as we navigate, attempt to better understand, and have fun with AI within academic libraries, empowering employees to navigate AI's complexities responsibly, creatively, and strategically.

AI-powered research tools are increasingly used by students and researchers, but are they fully aware of their limitations? Tools such as Elicit, Consensus, Scite.ai, Research Rabbit, Semantic Scholar and others are attracting researchers for their perceived ease of use, relevance, and ability to synthesize research. These tools are advertised on social media and LibGuides but with little discussion about their quality and cost.

Although most tools have free versions with limited features, the increased commercialization of AI research tools highlights the information privilege of researchers with more funding and presents barriers for researchers who cannot afford them, which may disproportionately affect researchers from marginalized groups. In this commercialized environment, startups are also less likely to provide information about their indexing and algorithms, considering them the “secret sauce” that gives them an edge over their competitors.

From a cursory review, these tools seem to be scraping open access repositories and excluding a large body of research behind paywalls. These tools also do not have features for researchers to access materials through their institutional access, further separating researchers from the library. Researchers should also consider how prioritizing productivity, embracing the publish or perish mindset, and over relying on AI tools for evidence synthesis may lead to a shallow understanding of the literature that affects the quality of their work.

This presentation will give a few tips on how librarians can discuss the strengths and weaknesses of AI tools with researchers to best support them in conducting comprehensive literature reviews.

3:15 - 3:30 p.m.

Libraries and librarians are vital leaders in the open education movement. It is increasingly common for them to support faculty as they transition courses to zero-cost course materials and create or adapt open educational resources (OER), among other things. Given the transformative nature of generative AI, librarians need to understand how these technologies fit into the OER landscape. This session will provide a brief overview of key opportunities and issues related to using generative AI for – and as – OER.

Artificial intelligence is rapidly changing the world, and libraries have a unique opportunity to shape our uses of generative AI intentionally. As an MSIS student and GRA in the implementation of AI for UT Libraries, I have been able to conduct research on how we can meld our organizational values with our technical capabilities and needs if we decide to implement an after-hours AI chatbot extension for our library chat service.

Ask a Librarian at UT Libraries is a valuable resource for students and faculty, providing an important connection between the library and campus life. An after-hours AI chatbot can extend the service to provide 24/7 support. In order to do so ethically, we must ensure that it benefits the stakeholders in such a project, including the library itself, the librarians, the students, the university staff, and all present and future library chat users.

This project involved a number of steps:

1. Analyzing chat logs to understand user needs and common queries
2. Looking to existing research on AI ethics and Human-AI Interaction as well as library codes of ethics to ensure responsible and appropriate development
3. Conducting and analyzing semi-structured interviews with Ask a Librarian staff to assess insights and concerns
4. Programmatically preparing existing knowledge sources on our website for ingress as knowledge for the chatbot
5. Analyzing the pros and cons of different chatbot creation platforms and creating a proof of concept

Throughout the project, AI played a crucial role in accomplishing more in a shorter time frame. By leveraging AI tools and techniques, I was able to streamline data analysis, automate knowledge preparation, and develop a more robust chatbot prototype.

In this presentation, I will discuss each step of the project, why it was important, and how I did it. I will also share how AI assisted in the process and discuss the possibilities for taking this project further.

The recent developments of generative artificial intelligence (AI) tools holds the potential to enhance all aspects of the teaching and learning process while also impacting the administrative and day-to-day operations of higher education institutions and libraries. In addition, the demand for college graduates to have high technical skills, including AI skills, are expected to rise exponentially. To ensure that our students are best prepared for a future-proof workforce, higher education institutions and academic libraries cannot shun AI. Instead, these organizations must balance the need to maximize these rapidly evolving technologies while also conscientiously addressing the ethical concerns they bring. In order to ensure the responsible and fair utilization of these tools, institutions and libraries must develop policies to encourage their ethical use. 

This presentation will provide sample artificial intelligence policy language from various higher education institutions and academic libraries. Topics covered will include the acceptable use of AI in the classroom, the role of faculty in making AI-related decisions, syllabus statements, AI use and detection, AI literacy, and library policies regarding AI. Participants will be able to compare and contrast policies to help them develop their own policies that work for their unique organization.

3:30 - 3:45 p.m.

Exploring the innovative use generative AI in revolutionizing curriculum mapping in information literacy instruction. Generative AI presents an opportunity to streamline the process of mapping information literacy objectives to curriculum, offers efficiency, accuracy, and adaptability.

The implementation of this process allows for more personalized learning and adaptive teaching methods. Paving the way for a more efficient and effective approach to improve the dreaded "one-shot."

This presentation will teach attendees how to use generative AI for legal research. It will discuss what tools are available for legal research, as well as the strengths and weaknesses of these tools. Particular focus will be on how to best prompt generative AI when conducting legal research and how to evaluate the results.

There must be a first look at data and the ethical aspect of its use as a key element for work and to what data refers. Data ethics builds on the foundations provided by information ethics and assesses data practices. An important question is raised what are the implications of data literacy for ethical librarianship? 

Libraries can provide the ability to think critically about data in different contexts and to reflect on the impact of different approaches to data and information. The AI Literacy Imperative is the assumption that today we have a critical need to understand and be able to use the key aspects of AI literacy. 

AI can bring many benefits to libraries, improving efficiency, facilitating information management, improving search and access to resources, and supporting research activities. When implementing AI in libraries, it is crucial to take care of ethical and security aspects, ensure data quality, and take into account user needs and requirements. In this way, AI can make a real difference in the efficiency and quality of the services that libraries provide to their users.

We can do that, we need to involve users and they need to have insight into how AI processes data and an understanding of how they arrive at certain results or conclusions. In this excellent place for the new role of libraries we need to be very careful when integrating generative tools because they don't just need data - they drive it. Data can be biased and unreliable, which comes not only from the data itself but also the design of algorithmic systems and how they are implemented (and who implements them). Ensuring that data will also be key to helping users to be aware of how much they can trust the data and the technologies and techniques they use.

3 - 3:45 p.m.

The rapid advancements in generative artificial intelligence (AI), exemplified by the rise of ChatGPT by Open AI, have significantly impacted the landscape of information literacy education. This presentation shares how Perplexity.ai can be leveraged to enhance students' information literacy skills when used in conjunction with the ACRL Framework for Information Literacy for Higher Education. We will focus on how to use Perplexity.ai to facilitate learning around "Research as Inquiry." Perplexity.ai, unlike ChatGPT, provides users with cited sources for the information it retrieves based on the user's questions and prompts. As a result, the tool utilizes a more transparent and accountable approach to information retrieval. 

While using Perplexity.ai, students:

• encounter the iterative research process.
• engage in verification of information and critical evaluation of sources.
• practice a strategic approach to navigate vast information sources.

Moreover, Perplexity.ai not only enables students to practice formulating research questions and evaluating sources, it also allows them to enact the very core of the the Frame: that ""research is iterative and depends upon asking increasingly complex or new questions whose answers in turn develop additional questions or lines of inquiry in any field.

This presentation highlights ideas to empower students to grow as information-literate researchers, using a generative AI tool, while upholding academic integrity. Rather than relying on piecemeal policies, this presentation advocates for a comprehensive integration of AI-based tools and the ACRL Framework.

Attendees will complete the session by breaking out into rooms to practice using Perplexity.ai in pairs or small groups. Example search topics will allow attendees to model the student-librarian interaction.

Both familiar and unfamiliar experiences arise from the transformative impact of artificial intelligence (AI) on medical librarianship. These cutting-edge technologies are revolutionizing how information is created, accessed, analyzed, and shared while also changing the nature of work in the field. One area where generative AI is especially promising is enhancing the quantity and quality of work performed by medical librarians.

However, the economic impact of generative AI is not uniform across the field. It can exacerbate racial and gender inequality in medical librarianship.  Furthermore, the relationship between the composition of health librarians and the probability of AI benefits or injury is yet to be investigated. By examining the current makeup of medical librarians and correlating it with future employment changes in this AI-dominated age, we can gain valuable insights into how AI will shape the growth of medical librarianship employment. This information can help us anticipate future developments and make informed decisions about how to best prepare for the changing landscape of health sciences librarianship. It's time to take action to ensure that the potential benefits of AI are shared equitably across all medical librarianship community members.

3:45–4 p.m.

This is a tentative schedule, subject to change.

Please note, all times are listed in Eastern Time on this agenda.

AI & Information Literacy

Sessions in this track explore the intersection of information literacy and AI, discussing strategies for teaching students to critically evaluate AI-generated content, navigate AI-shaped information landscapes, and understand the limitations and biases of AI systems. Presenters share innovative approaches to integrating AI into information literacy instruction, such as using ChatGPT to develop activities, comparing AI tools to track hallucinations, and examining the impact of input language on AI outputs. Attendees will gain practical ideas for promoting AI literacy alongside traditional information literacy skills.

Day 1: June 11

What is the role of librarians in helping to drive and support AI strategy on campus? This session will both invite attendees to explore this question, and discuss the creation of an AI Literacy Certificate Course being created by Babson College's Adjunct Lecturer & Digital Literacy Librarian. This certificate-style course will be offered to students who are interested in engaging with foundational concepts related to AI technologies and the social, ethical, and economic considerations involved in their application. By the end of the course, students will be able to articulate the responsibilities and expectations of someone living, learning, working, and creating in an AI-embedded world. The certificate’s completion can be featured as part of their professional/academic portfolios. This is an extremely transferable project. Digital Literacy education should be at the heart of academic libraries in the 21st century, and AI Literacy is becoming a critical component of Digital Literacy. The goal of this session is to inspire other academic libraries to create courses or learning objects related to AI Literacy, and take on an active role in AI strategy conversations and efforts in their communities.

Over the past several semesters, students in my undergraduate history courses have been given a 7-step research paper assignment designed to eliminate the use (or rather, misuse) of artificial intelligence.  In addition, this research paper assignment enabled them to better understand and utilize historical research processes.  However, over time I came to realize that my students were not fully benefiting from these series of exercises as they eliminated the possibility for additional learning and exploration. I became deeply aware that my teaching was being influenced by the fear of AI rather than the potential of it.  To address this concern, I revamped my historical research paper assignment so as to make AI a central component of it.  The initial results have shown great success and promise for future growth.  In this presentation, I would like to share my pre-AI and post-AI approaches while providing an examination of both.  I hope to make my audience aware of my reasoning for revamping this research paper assignment while detailing the steps that I took to build a viable AI appropriate alternative.  

Hey fellow instruction librarians! Get ready to embark on an exciting journey into the world of Artificial Intelligence (AI) and its potential to revolutionize library instruction. This conference session delves into the amazing possibilities of using generative AI to outline bibliographic library sessions. 

In this interactive gathering, we'll explore how AI tools can enhance our session planning and delivery. We'll dive into accessing AI tools, crafting session objectives and assessments, and developing comprehensive lesson plan outlines. We'll explore user-friendly AI tools designed specifically to make our session planning process smoother and more efficient. Discover techniques for clear, measurable goals and effective assessments. Unlock the magic of lesson outlines with AI-generated content, tailored to audience needs. 

This session is perfect for librarians, educators, and information professionals eager to harness the power of AI in their session planning and delivery. Bring your enthusiasm and curiosity, and let's embark on this AI-powered journey together!

Day 2: June 12

The launch of GPT-4 in March of 2023 led to a lot of buzz about artificial intelligence, much of it fearmongering and misinformation. With rapidly evolving, competing reports on the impact of AI (that it will replace - or assist? - workers, that it will exacerbate - or solve? - the climate crisis), patrons and librarians alike expressed concerns about exploring generative AI tools on their own. Due to the lack of trustworthy resources to help users navigate this rapidly evolving technology, generative AI emerged as an urgent site of both information literacy and skill development. 

Libraries and librarians are uniquely positioned to stem the flow of misinformation, alleviate the culture of fear, and to provide classes and tools that aid in skill development around new technologies. As passionate advocates for digital equity, we created a class that would render AI approachable and accessible. This biannual class addresses the basics of artificial intelligence, useful applications for life and work, its limitations and ethical implications, and discussions of the possible futures of AI technology. Through the removal of real and perceived barriers, we help users navigate the minefield of information about Generative AI, so patrons are empowered to draw their own conclusions and to use AI as they choose. 

This presentation will cover the process of researching, planning, instructing, and updating this biannual class. We will discuss how we used several Generative AI programs to familiarize ourselves with their operation. We will demonstrate AI's value as an instructional and research tool, detailing how Generative AI was used to help develop our lesson plan. We will share how we refresh our research to ensure we are dispelling misinformation, sharing accurate and up-to-date information, and demystifying artificial intelligence so that it is accessible to all.

In a world where the nuance of sustainability has become critically important, the intersection of Library instruction and AI utilization has made way for adventurous information seeking, analysis and use. Integrating sustainability across the curriculum involves getting faculty excited about meeting students where they are (on ChatGPT) while fostering information and AI literacy. Both of which may be topics that they themselves use, but don’t fit within their subject matter expertise. Librarians can provide instruction that not only benefits the students, but gives faculty a connection point that resonates with course level learning outcomes, deeply. This proposal outlines an interesting instruction model that has been useful with Librarians connecting sustainability concepts to the ACRL frameworks, with FILM students connecting the green production PEACH checklist to genre projects and with ENST Environmental Studies students who want to connect environmental impact to AI use. This interdisciplinary framework aims to foster environmentally sound, socially just, and economically feasible thinking, aligning with the core principles of sustainability, drawing out elements that take us all from literacy to competence. Case examples will lead attendees through three specific and detailed working examples used at Cal Poly Humboldt with faculty and students.

The prevailing sentiment among Humanities faculty on our campus is that Artificial Intelligence (AI) merely serves as a "plagiarism machine." This perception echoes the skepticism that greeted the internet in the 1990s and Wikipedia in the 2000s. However, just as with previous technological advancements, integration and critical engagement can lead to progress and enrichment. Banning AI doesn’t help our students think about their careers after college, or their life-long learning. Now is the time for exploration and experimentation, particularly in fields often considered traditional and conservative in light of new technologies. 

This presentation illustrates how generative AI can be used ethically and responsibly in the Humanities as a:

1. Research assistant
2. Summarizer
3. Data extractor
4. Text transformer
5. Tech support specialist
6. Visual Illustrator
7. Writing Assistant

This is information I share with teaching faculty as part of our discussions around information literacy instruction classes with the goal to demonstrate AI’s potential role in refining analysis, fostering creativity, and facilitating discovery. In two early success stories, I will share how an English lecturer used AI as a summarizer in her first-year composition class, and how a History professor employed AI in an upper-level class to analyze primary sources. 

This presentation will be useful to librarians across library sizes and types and will be made engaging and accessible using relatable examples, accessible presentation slides, clear visuals, and an accompanying handout (if possible).

By applying generative AI with careful analysis and ethical guidelines, faculty in Humanities fields will undoubtedly discover new research and scholarship opportunities. Early adopters will help to foster a harmonious relationship between this innovative technology and their enduring quest for knowledge in the humanities.

Day 3: June 13

This session aims to offer insights into the transformative power of AI in education, from teacher support to library innovation and mental health advocacy. This presentation will explore how AI tools are revolutionizing teaching practices and fostering innovation in school libraries. Magic School AI, KhanMigo, and Canva platforms are examples that can support teaching practice and will be discussed. We also discuss how Retrieval-Augmented Generation (RAG) provides ways to limit AI hallucinations and incorrect returns on queries. ISTE’s StretchAI will be used as an example.

GenAI can support teachers, from lesson planning and assessment to fostering meaningful engagement with students. We will discuss how AI can support new teachers and nurture the librarian-novice teacher relationship, thereby instilling resilience and limiting burn-out. We will also highlight innovative uses of AI in school libraries, emphasizing evaluation methods and AI literacy lessons that have successfully been implemented. GenAI has a role in promoting mental health awareness as well as the opportunity for timely interventions. We expand on the notion of mental health support using AI. Through interactive demonstrations and real-life examples, our presentation aims to inspire educators to embrace AI as a valuable ally in their teaching journey. 

We discuss how to implement guidance and policy, imperative as guardrails as well as supports for the use of GenAI. We will use specific examples to help frame this important aspect of GenAI use.

This presentation reframes the library one-shot as an opportunity to engage students, Faculty and community patrons by casting them in the role as co-creators. Using a multitude of Ai based web applications and games, attendees will get a broad overview of how to leverage these programs to teach critical thinking and literacy skills. A major challenge to chunking up Ai is its speed of adoption and ubiquity in contrast to its opaque construction. This becomes even more of a challenge when considering the digital divide and the intersectionality of information borders.

The goal of this modular presentation is to use Ai as a vehicle for discussing information search skills; data safety practices; the benefits, harms and risks of large language API’s, accessibility gains and deficit models; contextualizing Ai and other writing tools as extensions not replacements; centering writing as critical thinking and the importance of TASL attribution in light of the tenuous copyright status of computer assisted output. No previous coding, GAN or huggingface experience required!

The rapid advancements in generative artificial intelligence (AI), exemplified by the rise of ChatGPT by Open AI, have significantly impacted the landscape of information literacy education. This presentation shares how Perplexity.ai can be leveraged to enhance students' information literacy skills when used in conjunction with the ACRL Framework for Information Literacy for Higher Education. We will focus on how to use Perplexity.ai to facilitate learning around "Research as Inquiry." Perplexity.ai, unlike ChatGPT, provides users with cited sources for the information it retrieves based on the user's questions and prompts. As a result, the tool utilizes a more transparent and accountable approach to information retrieval. 

While using Perplexity.ai, students:

• encounter the iterative research process.
• engage in verification of information and critical evaluation of sources.
• practice a strategic approach to navigate vast information sources.

Moreover, Perplexity.ai not only enables students to practice formulating research questions and evaluating sources, it also allows them to enact the very core of the the Frame: that "research is iterative and depends upon asking increasingly complex or new questions whose answers in turn develop additional questions or lines of inquiry in any field.

This presentation highlights ideas to empower students to grow as information-literate researchers, using a generative AI tool, while upholding academic integrity. Rather than relying on piecemeal policies, this presentation advocates for a comprehensive integration of AI-based tools and the ACRL Framework.

Attendees will complete the session by breaking out into rooms to practice using Perplexity.ai in pairs or small groups. Example search topics will allow attendees to model the student-librarian interaction.

This is a tentative schedule, subject to change.

Please note, all times are listed in Eastern Time on this agenda.

AI Impact on Library Services

This track delves into the transformative impact of AI on various aspects of library services. Sessions cover topics such as developing conversational AI agents to enhance user interactions, leveraging AI for inclusive library experiences, and harnessing AI to improve engagement through gamification. Presenters also discuss the role of libraries in shaping AI strategy on campus and share case studies of successful AI implementations. Attendees will gain insights into how AI can be responsibly integrated into library workflows to improve efficiency, accessibility, and user experience.

Day 1: June 11

Given both the potential benefits and risks of generative AI, the need to educate students and instructors about AI tools is urgent. Librarians can leverage their existing roles in supporting information literacy in the instruction space to help foster the development of robust AI literacy. However, the sheer speed at which tools appear and evolve can turn this aspiration into a moving target.  

In this session, the presenters will first share a ChatGPT-based library instruction activity they developed to merge AI literacy with classroom objectives informed by the Association of College and Research Library’s Framework for Information Literacy in Higher Education using universal design for learning principals. They will emphasize the importance of developing students’ critical literacy skills such as discerning source material and evaluating the accuracy of generative text output. By incorporating group learning and hands-on activities, they will explain the importance of meeting the needs of learners with varying ChatGPT access and experience levels. Finally, they will reflect on how ChatGPT’s functionality has changed and will impact instructional endeavors by providing examples of potential future implementations.  

Participants will also engage in interactive brainstorming in the session to consider possible ways to incorporate generative AI tools into library instruction, including AI-based strategies to streamline the creation and updating of lessons and activities. They will more fully understand how to help students develop information literacy skills necessary to embrace emerging technology.

In today's rapidly evolving digital landscape, libraries play a pivotal role in fostering inclusivity and accessibility for patrons of all abilities. This presentation proposal aims to explore the transformative potential of artificial intelligence (AI) tools and robots in enhancing library services to meet the diverse needs of our communities.

From AI-driven virtual assistants delivering tailored support for visually impaired individuals to robotics aiding mobility, and accessibility within library spaces, participants will be introduced to a wide range of cutting-edge technologies that promise to redefine the accessibility of library experiences for patrons of all ages and abilities.

This presentation will showcase AI tools and robot innovations that have the potential to enhance library services, including integrating AI tools into makerspaces, children's programming, and specialized services for seniors. Additionally, the presentation will address some of the strategies for integrating these tools into libraries in a responsible way as we discuss the larger issues around AI such as ethics, copyright, affordability, accessibility, security, and user privacy.

By embracing AI for inclusivity, libraries have the unique opportunity not only to meet the diverse needs of their patrons but also to pioneer the creation of accessible and inviting environments for all community members. Please join us as we embark on a journey to explore the vast potential of AI in shaping a more inclusive future for libraries.

Retrieval Augmented Generation enables a rapidly expanding class of AI-powered research tools (e.g. Perplexity, Elicit, SciSpace) to conduct searches of articles and/or the broad Internet and then use the results of those searches to cite authentic sources and develop detailed answers. In this session we examine the processes by which these tools translate user-supplied prompts into search queries and then explore how we can use this knowledge to engineer our prompts to increase the relevance and quality of the retrieved results and, concomitantly, the supplied answer. We conclude with a discussion of how librarians can incorporate these methods into research consultations and how we can let our current and potential patrons know that we can help them make better use of these tools.

Day 2: June 12

The word salon is French, originally meaning "reception room." In 1800's France, the meaning grew to include a "gathering of elegant people" occurring regularly in such a room. Rather than focusing on building skills or producing deliverables, it encourages philosophical debate and engagement in current events and ideas as they expand.

In 2023, George Mason University Libraries initiated an internal summer salon series, inspired by historical science salons as forums for intellectual exchange. This series aimed to cultivate open dialogue among library staff regarding the impact of artificial intelligence (AI) tools in higher education and society. The salons encouraged philosophical discourse and engagement with contemporary issues, particularly surrounding technologies like ChatGPT.

Over three months, six scholarly discussions delved into the effects of AI tools on libraries, higher education, and society. Subject librarians facilitated these community-building sessions, both online and in-person. The success of the summer series led to its expansion into a campus-wide initiative for 2023-2024, covering diverse topics such as "AI in the Social Sciences and STEM," "AI in Business and Government," "AI, Ethics, and Academic Integrity," and "AI, Biases, and Prejudices," among others. Recognizing the need for further opportunities for discourse, the organizers transitioned to offering the salons in both in-person and Zoom-only formats. This shift toward a salon-style approach proved pivotal in fostering comprehensive conversations and involving the broader university community in exploring AI's interdisciplinary and intersectional impacts.

This presentation will focus on adopting a science salon format for discussions on ChatGPT and similar tools. It will outline the outcomes of the salon series and its potential applications in libraries of all sizes, addressing marketing strategies, assessment methods, and providing recommendations based on insights gained from these gatherings. Real-time accessible slides will be available for participants to download during the presentation.

Semantic Search (aka Vector Databases) have driven a massive increase in the widespread availability of effective search tools. However, the non-deterministic nature of these tools can make them unsuited for common academic tasks like systematic literature reviews. How can researchers take advantage of better search, while still having reproducible results? We'll explore tools like query expansion (SPLADE) and AI assisted screening, which can be powerful tools towards this end.  

Arizona State University (ASU) recently partnered with OpenAI to launch a Spring 2024 AI Innovation Challenge, where groups could propose various experiments to explore how generative AI is shaping the future of learning, research, and work in higher education. Those that were approved gained access to ChatGPT-4 licenses for approximately 8 weeks, experimenting with the system to encourage growth in knowledge, learning, and academic excellence. As AI becomes ubiquitous in higher education, librarians, who have unique takes on issues like authority, relevance, and accuracy, should have central roles in assessing these tools. ASU Library is participating in the AI Innovation Challenge to advance AI literacy skills and knowledge about AI’s uses, while also acknowledging its limitations across all libraries.

ASU accepted two ASU Library proposals, creating two teams to complete the projects. An evaluation framework project investigates whether we can develop a source evaluation framework (like CRAAP or SIFT) to assess text generated by ChatGPT. A personas project involves creating AI personas to simulate different patrons with various research skill levels to train new personnel for ASU Library’s “Ask a Librarian” chat service. In both projects, which are expected to wrap up in May, the teams are learning how best to use ChatGPT. 

This panel will reflect on the two projects’ goals and outcomes, including whether we can create an effective framework and whether AI can train new library employees. Discussion will offer insights into obstacles we encountered and critical issues that must be acknowledged when working with ChatGPT, like bias, diversity, hallucinations, “truthiness,” and data sourcing. Panelists will share their various experiences with ChatGPT and how they approached these issues and others. Furthermore, this panel will focus on AI’s influence through lenses of reference behavior, instruction, representation, and knowledge sharing. Lastly, if time, attendees can pose questions to panelists.

Day 3: June 13

At Minnesota State University Mankato (MNSU), we’ve undertaken several experiments and initiatives focused on Generative Artificial Intelligence. At the start of the fall semester, we collaborated with university Information Technology Services to present a professional development session for returning faculty through the MNSU Center for Excellence in Teaching & Learning on “5 Tips for Teaching with AI.” We also presented to librarians across the regional consortium, Minitex, on “The Library & Generative AI.” This presentation included several demonstrations. It was offered as an introduction to Generative AI focused on topics most relevant to librarians, including information literacy, as well as copyright and license-related concerns. Later in the fall, we offered a daylong “Experience Friday” workshop for area high school students on “AI at the MNSU Library.” The workshop included sections on basic prompting, responsible use, learning with AI, the future of work, and the dark side of AI. Throughout the fall, we experimented with using ChatBots to help with licensing as described in the NASIG Fall conference presentation, “AI as a License Review Assistant.” Finally, leading into the spring semester, we also experimented with using Dall-E to develop collections data illustrations as described in the SUNYLA conference presentation, “Novelty Visualizations of Collections Data: Real Impact or Comic Interlude?” At the GAIL virtual conference, we will review these initiatives and others to consider what has worked best and what we might not repeat. As we look to the future, we are asking ourselves, where should we focus our efforts?

My presentation will examine academic opinions regarding Generative AI, its potential uses, and potential ethical concerns regarding the tool. Attendees will also be able to use Generative AI tools in a constructive and effective manner.

In order to demonstrate the abilities of generative AI, attendees will participate via laptop in a number of activities (and discuss them in small and large groups), including: creating increasingly precise inquiries, using ChatGPT to create assessment tools using Canva AI image creator to create images for programming and/or personal use, examining Claude responses to learn about the AI’s syntax and habits, using ChatGPT to formulate responses to assignment prompts, and fact-checking Bing AI.

Most of the time will be involved with discussions, experiential training, and/or demonstrations. Attendees will be able to use generative AI tools or see them demonstrated in an effort to learn how to use them in a constructive manner. This training will be focused on how to use generative AI effectively in a wide variety of settings, demonstrating value to all types of peoplel

Through these experiences, librarians and staff will learn both about the subjects that their patrons want to research and how to effectively use technology for non-research purposes. They can apply the lessons learned through generative AI-enhanced programming as they augment their work products and tranings through the use of text generators, image diffusers, audio generators, and other tools. They can provide complex prompts and learn how the chatbot will react to certain phrases.

• Takeaway 1: -- Evaluate the potential of generative AI in creating library resources through personal experiences.​
• Takeaway 2: -- Understand the importance of librarians' attitudes towards technology in shaping students' ethical use of these tools.​
• Takeaway 3: -- Understand and apply the TCoP model of technology integration in the context of generative AI.​

Both familiar and unfamiliar experiences arise from the transformative impact of artificial intelligence (AI) on medical librarianship. These cutting-edge technologies are revolutionizing how information is created, accessed, analyzed, and shared while also changing the nature of work in the field. One area where generative AI is especially promising is enhancing the quantity and quality of work performed by medical librarians.

However, the economic impact of generative AI is not uniform across the field. It can exacerbate racial and gender inequality in medical librarianship.  Furthermore, the relationship between the composition of health librarians and the probability of AI benefits or injury is yet to be investigated. By examining the current makeup of medical librarians and correlating it with future employment changes in this AI-dominated age, we can gain valuable insights into how AI will shape the growth of medical librarianship employment. This information can help us anticipate future developments and make informed decisions about how to best prepare for the changing landscape of health sciences librarianship. It's time to take action to ensure that the potential benefits of AI are shared equitably across all medical librarianship community members.

This is a tentative schedule, subject to change.

Please note, all times are listed in Eastern Time on this agenda.

Ethical Considerations & Policy in AI 

Sessions in this track grapple with the complex ethical implications of AI adoption in libraries and higher education. Presenters explore issues such as data privacy, algorithmic bias, intellectual property rights, and the need for responsible AI governance. Discussions cover the development of AI policies, the impact of AI on accessibility and equity, and strategies for supporting diverse groups of users, such as PhD students. Attendees will gain a deeper understanding of the ethical challenges surrounding AI and learn best practices for promoting its responsible use in library contexts.

Day 1: June 11

Join me for this 45-minute discussion-based review of institutional policies created by libraries or their larger institutions on the use of Gen AI in teaching, learning, and research. During the first 15 minutes, I will share a selection of policies from 5-10 institutions, highlighting examples of commonalities and key differences, including how each policy addresses ethical aspects of using AI in the institution's context. The second 15 minutes will be breakout small group discussions of the example policies, policies participants are aware of or use at their own institutions or at institutions they’re curious about; and each group will have an online space to jot down notes and add links to policies or resources they discuss. The last 15 minutes will include a 1-2 minute report back from each group about useful aspects they found in the example policies or other policies discussed in their group, questions or concerns about policies discussed, examples of applying such policies at their institutions, examples of how to stay up to date with changes in Gen AI usage and make nimble adjustments to policies, or recommendations and comments for Gen AI policies going forward. We’ll conclude with a wrap up and links to the shared discussion documents for reference.

This session will examine the potential use cases for generative AI and explore the ethical considerations and potential impacts of the technology on the fabric of society. It will ultimately demonstrate that the harms inherent in such a technology and the unequal benefits it will provide to an already heavily unequal society should cause libraries - while instructing patrons on it - to reject it as antithetical to our profession and mission.

The Collingridge Dilemma posits that the impact of a given technology will not be understood until long after it has become too difficult to change.  If generative AI (GenAI) has not already reached this inflection point, it will soon.  Tracking the ethical implications of GenAI is challenging to library stakeholders because of its rapid technological advancement and proliferation.  This presentation will attempt to support librarians by providing a broad scan of ethical issues inherent in AI generated content, rather than focusing on one or two major issues.  Topics discussed will include data and algorithmic bias, plagiarism, misinformation, censorship, environmental impact, equity, and more exotic issues like cyberreligion.  Each ethical issue will include a description as well as documented examples.  By exploring with this environmental scan, we can more accurately strategize and frame the library profession's stance towards GenAI.  The presentation will conclude with a discussion on how to mitigate ethical concerns with GenAI, whether or not it has already passed the Collingridge inflection point.  This will also include a brief best practices primer on how to ethically engage with AI technology despite low user agency. 

Day 2: June 12

AI technology is a tool that will become commonplace in the near future. It is our responsibility to learn how to use this tool, and to use it wisely. My proposal is for a 45-minute seminar that covers the potential dangers and weaknesses of AI. It will be broken into two parts. 

The first part will deal with workplace safety and technological responsibility. It will cover topics such as data security, what information can safely be shared with AI systems, and the possible legal ramifications of misusing AI, even accidentally. AI is a product, and companies are creating excitement because they want you to use it, and they want you to buy it. Furthermore, if you are not the customer, then you are the product. Many companies are using AI to gather data, which can then be sold on the internet. We must therefore be careful to be responsible stewards of our patrons and organizations private information.

The second part will deal with social responsibility. AI is not an unbiased source of information. It can only be as reliable, and as unbiased, as the data that is fed into it. AI can be manipulated and controlled by individuals and organizations with private agenda’s. Furthermore, it can be influenced by cultural bias. AI is not a peer reviewed and reliable source of information. It can draw conclusions that are culturally biased, and lead to unequal outcomes for marginalized people. It can, and has, propagated dangerous and hurtful stereotypes about marginalized groups. Furthermore, AI has been known to make up fake data sets in order to fulfill requests. These digital “hallucinations” can skew data and misinform the public. All of these are dangers that we, as educators, must be aware of.

This presentation explores the evolving landscape of artificial intelligence (AI) in the context of educational accessibility and equity. It delves into the latest advancements in AI that aim to level the playing field for students of all backgrounds and abilities, while also discussing the ethical considerations and challenges that arise. The presentation aims to offer a balanced view of how AI can be harnessed for greater inclusivity in education, while also scrutinizing its limitations and potential for exacerbating inequalities.

In early spring, the presenters were invited to host a workshop for a group of thirty PhD students taking part a “Dissertation Bootcamp” organized by the university’s Graduate School. This presentation will describe the support we offered the PhD students during the workshop, as well as incorporate some of the questions the students asked during the session. We will highlight the guidance we provided related to relevant issues and concerns when using as a PhD student, including copyright/intellectual property, adhering to IRB protocols, reproducibility, bias, preparing for the defense, and establishing one’s own ethical boundaries. In addition to offering guidance on how to choose appropriate AI tools (and what to look out for), we will demonstrate several generative AI tools that could be most helpful for students in the dissertation-writing stage, including Perplexity AI, Keenious, Scite.AI, Research Rabbit, YouPro for Education, and Elicit. We also advice PhD students on how to enhance their research using AI-powered library resources. These tools can help them discover relevant materials, analyze data more efficiently, and gain deeper insights from their studies. Also, we will briefly discuss tools being developed by publishers such as Clarivate and Elsevier. We hope to use remaining time for questions and/or discussion.  

Day 3: June 13

Amid discussions of AI and machine learning tools and their impact on contexts of university teaching and library services, a common narrative has emerged: students are using AI tools extensively, a majority have acquired substantial expertise in AI tools, and the rest of the academic community is falling behind and must strive to catch up. 

But to what extent do these premises stand up to scrutiny? We sought the perspectives of students themselves via a survey whose findings challenge these prevailing assumptions. This presentation will discuss the survey results and their implications, highlighting the need to embrace a more nuanced approach to AI adoption in instruction and in the provision of library services. We conclude that we need to carefully assess assumptions about students’ AI proficiency and leverage librarians’ unique position to solicit student feedback about use of AI tools; that the inclusion of student perspectives in shaping AI adoption is vital; and that students are not uncritical adopters of AI and have nuanced insights and critiques that are evident in their choices about when and how to deploy AI tools to assist learning.

Lightning Round 1: 15-Minute Presentations (2 p.m. - 2:45 p.m.)

The advent of new technology has always heralded the perception that the demise of librarianship is imminent. Recent significant waves of technological change that come to mind include the transition to online “card” catalogs, the implementation of the World Wide Web, databases delivering online content,  and now generative AI. With each technological change, librarians have adapted and thus moved the profession forward keeping libraries at the heart of their communities. Generative AI has the potential to enrich our professional frameworks, specifically around Information Literacy and Scholarship as Conversation. It is offering new ways for stakeholders across campuses to interact with information, expand their perspectives, and contribute their voices to the ongoing scholarly discourse. Once again librarians should be at the center of this conversation as we help to structure how we focus our attention on generative AI.

When a new technology seems set to disrupt the information landscape, affecting libraries no less than the field of education broadly, it is understandable to feel as though the disruption puts everything we know into question. The fact is, libraries have weathered disruption after disruption in the last 40 years, from computers to the internet, social media and the ability of the citizen scientist to gain prominence over scholars backed by tenure and peer review. We have tools with which to approach the disruption of generative AI and its impact on library services and higher education. One such tool is the theoretical framework presented by Ranganathan’s Five Laws of Library Science (1931). With small shifts in language, we can update this framework to apply to modern information challenges in a way that preserves the core values that ground our practice. Guiding our users to the right resource for them, whether they are incoming freshman or tenured faculty, is extremely dependent on the context in which the resource is used and the end product that is envisioned. In order to uphold the rigor of academic publishing, librarians need to thoroughly understand emergent technologies and ground them into our foundational theory to prevent the academy from perishing. 

Mr. Mercer’s 15-minute presentation will explore the intersection between AI literacy, intellectual property (IP), and libraries. The session will discuss AI application in libraries and the information organization. Topics discussed will include the user experience, user interface, and neural networks and how they comprise the backbone of AI application user engagement front to backend. In doing so, the speaker will explain the implications between the user, the constraints of IP and its application / implementation as they apply to libraries and resource-rich data depositories. The session will conclude with a short discussion of (the above-mentioned) as it relates to diversity with an emphasis on people with disabilities (close captions provided), including the neurodivergent.

Lightning Round 2: 15-Minute Presentations (3 p.m. - 3:45 p.m.)

AI-powered research tools are increasingly used by students and researchers, but are they fully aware of their limitations? Tools such as Elicit, Consensus, Scite.ai, Research Rabbit, Semantic Scholar and others are attracting researchers for their perceived ease of use, relevance, and ability to synthesize research. These tools are advertised on social media and LibGuides but with little discussion about their quality and cost. Although most tools have free versions with limited features, the increased commercialization of AI research tools highlights the information privilege of researchers with more funding and presents barriers for researchers who cannot afford them, which may disproportionately affect researchers from marginalized groups. In this commercialized environment, startups are also less likely to provide information about their indexing and algorithms, considering them the “secret sauce” that gives them an edge over their competitors. From a cursory review, these tools seem to be scraping open access repositories and excluding a large body of research behind paywalls. These tools also do not have features for researchers to access materials through their institutional access, further separating researchers from the library. Researchers should also consider how prioritizing productivity, embracing the publish or perish mindset, and over relying on AI tools for evidence synthesis may lead to a shallow understanding of the literature that affects the quality of their work. This presentation will give a few tips on how librarians can discuss the strengths and weaknesses of AI tools with researchers to best support them in conducting comprehensive literature reviews.

The recent developments of generative artificial intelligence (AI) tools holds the potential to enhance all aspects of the teaching and learning process while also impacting the administrative and day-to-day operations of higher education institutions and libraries. In addition, the demand for college graduates to have high technical skills, including AI skills, are expected to rise exponentially. To ensure that our students are best prepared for a future-proof workforce, higher education institutions and academic libraries cannot shun AI. Instead, these organizations must balance the need to maximize these rapidly evolving technologies while also conscientiously addressing the ethical concerns they bring. In order to ensure the responsible and fair utilization of these tools, institutions and libraries must develop policies to encourage their ethical use. 

This presentation will provide sample artificial intelligence policy language from various higher education institutions and academic libraries. Topics covered will include the acceptable use of AI in the classroom, the role of faculty in making AI-related decisions, syllabus statements, AI use and detection, AI literacy, and library policies regarding AI. Participants will be able to compare and contrast policies to help them develop their own policies that work for their unique organization.

There must be a first look at data and the ethical aspect of its use as a key element for work and to what data refers. Data ethics builds on the foundations provided by information ethics and assesses data practices. An important question is raised what are the implications of data literacy for ethical librarianship? 

Libraries can provide the ability to think critically about data in different contexts and to reflect on the impact of different approaches to data and information. The AI Literacy Imperative is the assumption that today we have a critical need to understand and be able to use the key aspects of AI literacy. 

AI can bring many benefits to libraries, improving efficiency, facilitating information management, improving search and access to resources, and supporting research activities. When implementing AI in libraries, it is crucial to take care of ethical and security aspects, ensure data quality, and take into account user needs and requirements. In this way, AI can make a real difference in the efficiency and quality of the services that libraries provide to their users.

We can do that, we need to involve users and they need to have insight into how AI processes data and an understanding of how they arrive at certain results or conclusions. In this excellent place for the new role of libraries we need to be very careful when integrating generative tools because they don't just need data - they drive it. Data can be biased and unreliable, which comes not only from the data itself but also the design of algorithmic systems and how they are implemented (and who implements them). Ensuring that data will also be key to helping users to be aware of how much they can trust the data and the technologies and techniques they use.

This is a tentative schedule, subject to change.

Please note, all times are listed in Eastern Time on this agenda.

AI Implementation in Libraries

This track offers practical guidance and real-world examples of AI implementation in libraries. Sessions cover topics such as leveraging library expertise to inform institutional AI approaches, crafting effective prompts for AI-generated content, and using AI to enhance library instruction and research support. Presenters share their experiences with developing AI-powered chatbots, incorporating AI-generated images into library materials, and establishing staff AI interest groups. Attendees will leave with concrete strategies and considerations for successfully implementing AI solutions in their own libraries.

Day 1: June 11

When National Louis University began exploring the capabilities of generative AI and its potential role in the college classroom, University Library faculty harnessed their expertise in information literacy to bring key themes and frameworks in academic integrity education to the forefront of the discussion. By facilitating student focus groups and collaborating with stakeholders across campus, including the Center for Teaching & Learning Excellence, librarians helped the university develop an integrity-minded approach to AI policy and programming that centered the student perspective while also meeting faculty where they were at: from enthusiastic early adopters of AI to highly cautious beginners experiencing discomfort, fear, and uncertainty about AI tools in higher education. As a result, the University Library has established itself as a leading support for faculty in developing their own approaches to AI in the classroom and continues to help shape academic policies at the course, program, and university level.

Libraries have long served as hubs of interaction and innovation as they readily incorporate the latest technologies into their services, outreach, and even marketing. Generative AI serves as yet another technological tool to continue to enhance library services. Join us as we delve into the dynamic intersection of gamification and generative AI within academic libraries. This session will explore innovative strategies for leveraging AI to enhance library engagement through the gamification of information literacy instruction, outreach initiatives, and library marketing.  Attendees will leave the session with concrete examples of how to incorporate AI and gamification into many of their daily library duties such as teaching, designing outreach initiatives, and creating library marketing strategies. 

In early 2023, Zayed University Library (United Arab Emirates) began developing Aisha, an AI-powered chatbot using OpenAI's ChatGPT, Python, a vector database, and the LangChain framework. This was a pioneering project in academic library services, creating a chatbot that can communicate in naturally in multiple languages and provide users with enhanced access to the library's resources. Aisha has since been integrated with many internal and external data sources such as Google, Wikipedia, the library catalog and the institutional repository, to offer additional support. The project goes on and we are currently evaluating Aisha's effectiveness with various large language models, such as GPT-4, Google Gemini, Llama 2 and Anthropic Claude, to understand their applicability in the library setting. This presentation will cover the journey of developing Aisha, including the challenges we encountered, and our insights into different large language models in the library context. The presentation will explore how conversational AI can make library services more personal and efficient, potentially redefining library interactions globally. Through our ongoing efforts, we hope to offer valuable insights to other libraries considering similar technological advancements. Conversational AI agents like Aisha have a lot to offer and they could potentially become the future interfaces to all library services.

Day 2: June 12

Everyone’s talking about AI! It’s on the news, featured in webinars and conferences, and popping up at the reference desk. But for many in the library profession it still has not actually hit our desktops or workflows. In this session, we will provide guidance and tools to start developing your own AI literacy by identifying common terminology and useful resources, providing a brief overview of the current landscape, explaining how tools like ChatGPT work (and why they are so popular), and offering a framework to support your own learning. Attendees will build their confidence and conversational fluency around AI, enhancing their ability to evaluate AI products and database integrations and support library users as they explore and leverage AI tools for their own information needs. AI is for everyone, make it work for you!

Generative artificial intelligence (GenAI) tools such as ChatGPT have proliferated throughout the educational landscape faster than many educators can adapt. These emerging trends have made it essential for library users to hone their information literacy skills.  In response, the presenters have developed a set of dynamic lessons focused on the intersection between GenAI and information literacy as a means to develop user skills.  This session was designed in a ‘plug and play’ format to account for multiple instructional settings, relevant frameworks, and pedagogical methods.  The resulting lesson includes three active learning activities, tied to outcomes from ACRL’s information literacy framework.  The module was shared via a staff intranet to allow other library staff members to utilize and adapt these activities without needing extensive GenAI expertise.  This presentation will discuss the development, philosophy, and implementation of the lesson module, with emphasis on encouraging the wise and productive use of GenAI.

This presentation will appeal to library workers who are interested in incorporating Artificial Intelligence (AI) generated images into their work.  Much has been written about the limitations of images produced using AI. We will talk about how we successfully – or mostly successfully-- incorporated AI images into a professional presentation that covers a sensitive topic.  That presentation (forthcoming) focused on how larger-body individuals navigate situations and spaces in libraries and on university campuses.  We decided to incorporate AI images in the presentation because it was less intrusive than using actual photographs of large-bodied people.  In today’s presentation, we will walk participants through our approaches to generating these images using Dall-E.  Many of the generated images were unacceptable to us because they were disrespectful (cartoonish and rooted in stereotypes), but with persistence, we were able to generate usable images for most of the situations we wanted to depict.  In this session, we will talk about prompts that seemed to work best and those that didn’t work well.  We will discuss AI ethics, fair use, and biases. Our presentation will be useful for librarians who are interested in incorporating AI images into instruction materials, library publications, library web pages, and outreach activities.

Day 3: June 13

Discover how strategic prompt design can help you harness the power of generative artificial intelligence (AI) in your library.  Through a series of examples, the presenters will demonstrate the impact that well-crafted prompts can have on the quality and relevance of AI-generated outputs.  Attendees will learn how to create prompts for successful AI outputs while mitigating common AI issues of bias and hallucinations.  Real-world examples will include how to curate legitimate reading lists, craft MARC records, enrich metadata, and generate inclusive images.  Unlock your library’s full AI potential to enhance user experiences and patron engagement.

This workshop will introduce attendees to the use of Large Language Models for Colloquial Data Science. Colloquial Data Science is the use of natural language prompts to solve data science problems. While many in libraries have at least a passing familiarity with basic statistics, such as measures of central tendency (mean, median, and mode) and variation (standard deviation), fewer librarians or library staff are comfortable with programming. Large Language Models such as Anthrophic’s Claude, OpenAI’s ChatGPT, and Google’s Gemini excel at code generation. The workshop will cover the basics of using an LLM for performing exploratory data analysis—including generating summary statistics and visualizing various variables—of a dataset.

Lightning Round 1: 15-Minute Presentations (3 p.m. - 3:45 p.m.)

Learn how one academic library is taking a transformational approach to exploring the possibilities with artificial intelligence (AI) through the creation of an AI Interest Group. With AI rapidly reshaping the library profession and higher education, our AI Interest Group aims to equip all library employees with the knowledge, skills, confidence, and mindset to embrace AI responsibly and consider leveraging its potential in their library work. 

Since October 2023, the AI Interest Group has served as an inclusive forum that has brought together colleagues from roles across all library departments and with varying levels of familiarity with AI. Topics have included generative AI, social and emotional intelligence, bias and fairness, and the considerations involved in the ethical governance, policymaking, and regulation of AI. We even delved into discussions about the ownership, authorship, and attribution of AI-generated content. 

We’ll share how we gathered and prioritized ideas and subsequently offered programming to learn from one another through thought-provoking discussions, webinar watch-parties, interactive demos of specific AI tools, guest speakers from different disciplines on campus, and sharing experiences of using AI in our work.  

We’ll also highlight key considerations, including challenges, successful outcomes, and insights about launching and sustaining an engaging AI Interest Group that maintains high engagement each session. Attendees will leave with ideas, considerations, and a roadmap for establishing a similar grassroots initiative. 

Join us as we navigate, attempt to better understand, and have fun with AI within academic libraries, empowering employees to navigate AI's complexities responsibly, creatively, and strategically.

Artificial intelligence is rapidly changing the world, and libraries have a unique opportunity to shape our uses of generative AI intentionally. As an MSIS student and GRA in the implementation of AI for UT Libraries, I have been able to conduct research on how we can meld our organizational values with our technical capabilities and needs if we decide to implement an after-hours AI chatbot extension for our library chat service.

Ask a Librarian at UT Libraries is a valuable resource for students and faculty, providing an important connection between the library and campus life. An after-hours AI chatbot can extend the service to provide 24/7 support. In order to do so ethically, we must ensure that it benefits the stakeholders in such a project, including the library itself, the librarians, the students, the university staff, and all present and future library chat users.

This project involved a number of steps:

1. Analyzing chat logs to understand user needs and common queries
2. Looking to existing research on AI ethics and Human-AI Interaction as well as library codes of ethics to ensure responsible and appropriate development
3. Conducting and analyzing semi-structured interviews with Ask a Librarian staff to assess insights and concerns
4. Programmatically preparing existing knowledge sources on our website for ingress as knowledge for the chatbot
5. Analyzing the pros and cons of different chatbot creation platforms and creating a proof of concept

Throughout the project, AI played a crucial role in accomplishing more in a shorter timeframe. By leveraging AI tools and techniques, I was able to streamline data analysis, automate knowledge preparation, and develop a more robust chatbot prototype.

In this presentation, I will discuss each step of the project, why it was important, and how I did it. I will also share how AI assisted in the process and discuss the possibilities for taking this project further.

This presentation will teach attendees how to use generative AI for legal research. It will discuss what tools are available for legal research, as well as the strengths and weaknesses of these tools. Particular focus will be on how to best prompt generative AI when conducting legal research and how to evaluate the results.