Exploring Librarians’ Experiences with Generative AI: My Article in Library Trends

My latest article, “Use of Generative AI in Aiding Daily Professional Tasks: A Survey of Librarians’ Experiences,” has been published in Library Trends 73(3). This issue is the first of a two-part special theme: “Generative AI and Libraries: Applications and Ethics, Part I.”

As generative AI tools like ChatGPT, Claude, and Gemini become increasingly integrated into our daily lives, librarians across the globe are navigating both opportunities and challenges. This issue of Library Trends dives into that evolving landscape, providing a timely, thoughtful examination of how libraries are adapting to and making sense of generative AI.

AI is transforming how we interact with information, deliver services, and conduct research. Yet, it also raises important ethical questions about bias, authorship, intellectual freedom, and more. My article contributes to this conversation by offering empirical insights into how librarians are currently using generative AI to support their work. Through a national survey, I explored:

  • How librarians are using generative AI in tasks such as writing, planning programs, generating summaries, and more
  • Their motivations, expectations, and experiences
  • The benefits they perceive, including time-saving and enhanced creativity
  • Concerns and barriers, such as ethical implications, institutional policies, and lack of training

What stood out most to me in this research was the overall curiosity and cautious optimism among librarians. Many are experimenting with AI tools to streamline workflows and increase productivity, while also expressing a desire for clearer guidance, ongoing training, and thoughtful policies that center professional values. This article is just one step in a broader journey of understanding how our profession can thoughtfully harness technology without compromising the principles that define us.

If you’re interested in the intersection of AI and libraries, from practical applications to deeper ethical debates, I highly recommend reading this special issue of Library Trends. It features a wide range of voices and perspectives.

AI Literacy in Higher Education: The Role of Academic Libraries

AI literacy is defined as the ability to recognize, understand, use, and critically evaluate artificial intelligence (AI) technologies and their impacts. As AI permeates various industries, AI-literate graduates will have a competitive edge in the job market. AI literacy cultivates critical thinking, enabling students to comprehend AI’s limitations, ethical implications, and potential biases. AI literacy is a fundamental skill for navigating the AI era, allowing students to harness its benefits while mitigating risks.

Academic libraries can play a pivotal role in enhancing college students’ AI literacy. In the United States, libraries have curated AI-focused resources, developed workshops to demystify AI concepts, ethical implications, and applications. Furthermore, librarians can collaborate with faculty to integrate AI literacy into curricula, fostering interdisciplinary learning. 

Recently I conducted a pilot study to analyze libraries’ efforts in supporting AI literacy education via a content analysis study examing US libraries’ LibGuides on AI literacy. LibGuides are the most frequently used systems for US librarians to create user-friendly online guides or mini-websites on specific topics. This study specifically focused on the content and themes represented in AI Literacy LibGuides from a sample of ten academic libraries, including five research universities, three teaching universities, and two junior colleges.

The content analysis revealed a strong emphasis on the ethical use of AI across all sampled libraries, highlighting the significance of AI and academic integrity. Libraries were unanimous (100%) in their inclusion of guidelines for ethically disclosing AI use, understanding and utilizing AI detectors, citing AI-generated content, and recognizing AI’s impact on academic publishing and copyright. This uniform presence underscores the academic community’s emphasis on responsible AI usage and transparency in scholarly work.

Another widely covered theme was AI fundamentals, present in 90% of the libraries sampled. These sections included basic explanations of AI concepts, terminology, and the underlying technologies. This focus on foundational knowledge reflects a commitment to providing all library users, regardless of their prior AI knowledge, with a solid grounding in AI concepts.

Approximately 70% of the sampled libraries included sections defining AI literacy, discussing critical thinking and evaluation skills for AI-generated information, and emphasizing effective use of AI tools. These topics are central to helping students discern AI-generated content’s reliability and accuracy and providing guidance on using AI tools for academic purposes. This category also covered specific guidance on constructing effective AI prompts, ensuring that students and faculty could utilize AI tools to their fullest potential.

College-specific AI policies and resources, such as training workshops, also appeared in 70% of the LibGuides. This inclusion demonstrates how institutions tailor their AI literacy support to the unique needs of their communities. Moreover, 60% of the sampled libraries included AI-related guidance and resources specifically for instructors, acknowledging the importance of equipping educators to guide students effectively in AI literacy.

A smaller portion of the guides presented a balanced view of AI, covering both its benefits and limitations (40%). Only 20% of libraries highlighted environmental and ethical concerns associated with AI, such as environmental impact, labor exploitation, privacy issues, and risks of AI “hallucinations” (instances where AI generates false or misleading information). Similarly, only 20% of the libraries explicitly mentioned their own library policies or resources regarding AI, which suggests that formal AI policies may still be evolving within academic library settings.

Findings from the study highlight the priorities and evolving areas within academic libraries’ AI literacy efforts. While ethical use and foundational understanding dominate the content in these LibGuides, there is variability in how libraries address AI’s broader societal implications and institutional policies. As AI technology and its role in academia continue to develop, academic libraries may play an increasingly crucial role in equipping students and faculty with the critical skills and ethical understanding needed to navigate an AI-rich academic landscape.

Using ChatGPT in Research Design

In the past two weeks, we completed the research methods training portion of 2023 IRDL Online. IRDL is a continuing education program for academic and research librarians designed to create a growing community of confident librarian-researchers. The research methods training portion was provided in the form of an online summer research workshop, where I was an instructor. It’s incredibly energizing and exciting to talk to a group of research-minded librarians about conducting research to improve professional practice. One theme that arose from our discussions was the use of AI in research design. One of the librarian scholars gave it a try and asked ChatGPT to write survey questions for a research scenario.

Here’s her prompt:  I need to write a survey to find out how academic librarians engage in professional development. ACRL will use the survey findings to develop the Guidelines for Professional Development for Academic Librarians, which they believe will provide guidance for academic librarians as they explore professional development opportunities and make professional development plans. Can you write 4 questions for this survey?

ChatGPT’s response:
Certainly! Here are four sample questions for your survey on how academic librarians engage in professional development:

1. How frequently do you engage in professional development activities related to your role as an academic librarian?
• Rarely or never
• Occasionally (1-2 times per year)
• Regularly (3-5 times per year)
• Frequently (more than 5 times per year)

2. What types of professional development activities do you find most beneficial in enhancing your skills and knowledge as an academic librarian? (Select all that apply)
• Workshops and conferences
• Webinars and online courses
• Professional association memberships and events
• Formal education (e.g., degree programs)
• Informal learning (e.g., self-study, reading journals)
• Mentoring and coaching programs
• Other (please specify)

3. How do you typically seek out information about professional development opportunities? (Select all that apply)
• Internal communication channels (e.g., email, intranet)
• External websites and online resources
• Professional networks and colleagues
• Professional association publications and newsletters
• Supervisors and managers
• Other (please specify)

4. What factors influence your decision to participate in a professional development activity? (Select all that apply)
• Relevance to current job responsibilities and career goals
• Availability of funding or financial support
• Recommendations from colleagues or supervisors
• Reputation and credibility of the provider or organization
• Flexibility in terms of time and scheduling
• Recognition or credentialing opportunities
• Other (please specify)

Looking at these survey questions, I think ChatGPT did a nice job in operationalizing the key concepts in the research question based on the research goal. There’s room for improvement in wording survey questions that uphold the basic rules for survey design such as being concise and unambiguous, avoiding leading questions or double barreled questions, and making sure response categories are exhaustive and mutually exclusive. Still, this can be a good start if we are get stuck in our instrument design and don’t know where to begin.

There’s been discussion about using AI in manuscript writing such as this Science article “As scientists explore AI-written text, journals hammer out policies“. Looks like we will need more discussion about using AI in assisting the research process such as designing survey questionnaire, interview guide or a content analysis codebook.

Impact of the COVID-19 Pandemic on Chinese College Students’ Information Behavior and Library Needs

Last year, my research collaborators and I conducted a study to examine how the pandemic had impacted Chinese college students’ use of the library. The study took place in April and May 2020. At that time, Chinese college campus had been shut off and all the teaching and learning had been moved online for two months. We collected data using a combination of journaling and in-depth interviews, focusing on the library use and library needs of Chinese college students during the COVID-19 pandemic.

Our findings indicate that students generally lack awareness and understanding of the library’s online resources and would like the library to provide support in ways that could help them enhance productivity and lessen anxiety in the new reality of learning. Libraries should consider making more outreach efforts and offering programs and events to ameliorate isolation and improve students’ sense of community. We hope that our study can provide academic libraries with a nuanced view of user needs and thus help them make informed decisions to serve their campus communities during the unprecedented health crisis. Through the exploration and documentation of college students’ library use and needs, we also hope to document this critical historical event for the library community.

Now our study has been published in the Library Quarterly. Here’s the citation:

Shi, Y., Li, C. & Luo, L. (2021). Impact of the covid-19 pandemic on Chinese college students’ information behavior and library needs: A qualitative study. Library Quarterly, 91(2), 150-171.

I’m really glad that we were able to contribute to the library and information science (LIS) literature with regards to the pandemic. When years later, the pandemic is finally behind us, I think it will be quite interesting to conduct a content analysis of all the pandemic-related studies/articles in peer reviewed LIS journals to capture and document LIS researchers’ collection contributions.

Ideas for the COVID-19 Pandemic Related Research in Librarianship

Earlier this week I had an opportunity to talk to librarians at the University Library at the University of California at San Diego about the trends and methods of research in academic librarianship. They asked wonderful questions, and one of them was about the kinds of research that librarians may consider doing during the pandemic. The COVID-19 pandemic has certainly disrupted our lives in significant ways, but there may also be new research opportunities arising from it. Conducting research related to the pandemic can help library professionals better understand our user needs and provide more meaningful and effective responses to this public health crisis.

A few months ago when the pandemic just broke out, the Journal of the American Society for Information Science and Technology published a commentary titled “Global health crises are also information crises: A call to action” (Xie et al ,2020)“. In this article, a group of information researchers discussed specific things that information scientists can do to “help individuals and society as a whole survive global health crises like COVID‐19, deal with the aftermath, and be better prepared for the next crisis”. They recommended the following research directions:

  • Misinformation/disinformation particularly during global health crises
  • Health literacy—including eHealth literacy
  • Information behavior during lock downs
  • Vulnerable populations—a case for accessible and usable solutions
  • Information dissemination, sharing, and integration among multiple forms of digital data
  • eHealth tools
  • Predictive methods
  • Digital archiving
  • Ethical considerations

I think it would be helpful for library researchers to put out a similar “call to action”, encouraging librarians to investigate research topics related to the pandemic as well. Some of the potential topics may include:

  • Usage of library services during the pandemic, how it changes from before – especially the use of online library resources and services
  • Library needs of users during the pandemic – e.g. for academic libraries, how do their students and faculty would like the library to provide support to assist them as they study from home? for special user populations, especially those suffering from the digital divide, what can libraries to ensure equity when providing services during the closure of physical library locations?
  • Librarians’ well-being – what are librarians’ health and safety concerns with regards to working during the pandemic if they have to return to work?
  • How librarians can help address misinformation related to the pandemic – what can the library do to better help library users become more critical consumers of information and avoid being victimized by misinformation?
  • Libraries’ response to the pandemic – did the library have a crisis management/communication plan, how did the library make decisions on their responses to this crisis?

Information behavior during the “Infodemic” – Ideas for academic librarians

This COVID-19 pandemic has unleashed an information crisis. The World Health Organization (WHO) announced that the society not only needs to fight the pandemic, but also the “infodemic,” which it defined as “an overabundance of information—some accurate and some not—that makes it hard for people to find trustworthy sources and reliable guidance when they need it“. This “infodemic” could add another layer of challenge as college students strive to navigate the turbulent time.

College students are likely to fall prey to the spiraling misinformation during the infodemic. Standford researcher Sam Wineburg and his team have repeatedly found that college students were not well-equipped with the ability to critically evaluate online information, and they struggled to effectively evaluate online claims, sources, and evidence. Their research indicated that college students could be deceived by easily manipulated features of websites, such as official-looking logos and domain names, and they tend to read a webpage vertically, staying within a website to evaluate its reliability. In the meantime, high production values, links to reputable news organizations and polished “About” pages were likely to sway students into trusting the contents of the site without much skepticism.

Academic libraries have been increasingly involved in helping college students battle misinformation. Here are a couple of recent articles that explored academic librarians’ efforts on this front: “Academic library guides for tackling fake news: A content analysis” by Sook Lim; and “News Credibility: Adapting and Testing a Source Evaluation Assessment in Journalism” by Piotr S. Bobkowski and Karna Younger.

Yesterday, I read about the following study on TIME magazine, and thought this might be something that academic librarians may find useful as they guide students’ information behavior during the “infodemic”.

A new paper in Psychological Science explores not only why people believe Internet falsehoods but also how to help them become more discerning about what they share. One of the leading reasons misinformation about COVID-19 gains traction is that it’s a topic that scares the daylights out of us. The more emotional valence something we read online has, the likelier we are to pass it on.

That’s in keeping with earlier research out of MIT, published in 2018, showing that fake news spreads faster on Twitter than does the truth. The reason, the researchers in that study wrote, was that lies are “more novel than true news … [eliciting] fear, disgust and surprise,” just the things that give sharing its zing.

Political leanings also influence sharing. A 2019 Science study, from researchers at Northeastern and elsewhere, showed that neither the left nor the right has a monopoly on sharing fake news; mostly people are just choosing content that fits their ideologies.

To dig deeper still into sharing decisions, Rand and colleagues developed a two-part study. In the first, they assembled a sample group of 853 adults and first asked them to take a pair of tests. One, known as the Cognitive Reflection Test (CRT), measures basic reasoning processes. The other measured basic science knowledge. The sample pool was then divided in half. Both halves were shown the same series of 30 headlines—15 false and 15 true—about COVID-19, but they were instructed to do two different things. One group was asked to determine the accuracy of the headlines. The other was asked if they would be inclined to share the headlines.

The results were striking. The first group correctly identified the truth or falsehood of two-thirds of the headlines. The second group—freed from having to consider accuracy—reported that they would share half the headlines, equally divided between true ones and false ones.

The study did find that people who scored higher on the CRT and basic science tests were a little less indiscriminate. The solution, clearly, is not to force people to pass a reasoning test before they’re admitted online. The second part of the study provided a better answer.

For that portion, a different sample group of 856 adults was again divided in two and again shown the same set of headlines. This time, neither group was asked to determine the accuracy of the headlines; instead, both were asked only if they would share them. But there was still a difference: one group was first shown one of four non- COVID-19-related headlines and asked to determine whether it was true or false. That priming—asking the subjects to engage their critical faculties—made a big difference: the primed group was one-third as likely as the unprimed one to share a false headline.

Thinking critically—especially about the truth of a headline—reduces the spread of fake news.”

Assessing Scholarly Communication Services: A National Forum in May 2020

I’m a member of a great project team that’s working to put together the National Forum of the Assessment of Scholarly Communications Programs. The project is funded by IMLS and led by wonderful librarians from Sacramento State University and San Jose State University.

The forum will be held online via Zoom on May 4th and 5th. Attendees of the forum will include experts from library assessment that will present and lead discussions on how existing assessment techniques can be implemented for scholarly communication services. The forum will result in a report with recommendations for standards and a comprehensive set of best practices in assessing the range of services that comprise scholarly communication program.

My role in this grant project is to provide methodological expertise in the data gathering and analysis. I have been closely involved in designing and implementing the data collection instruments to gather input from librarians and campus stakeholders that will inform the project goal. We used the Research Lifecycle from the University of Central Florida to frame our data collection instruments, and the data has truly opened my eyes to the wide variety of practices in scholarly communication programs and their role in the campus research enterprise. Can’t wait to hear more from librarians and campus stakeholders at the Forum!

Image result for research lifecycle at university of central florida

Information literacy and critical thinking

Recently I read a book titled “A Field Guide to Lies: Critical Thinking in the Information Age ” by Daniel J. Levitin. It has some really practical implications for information literacy instruction. I particularly enjoyed how the author illustrated the deceptive ways that statistics can be manipulated and how he advocated “Bayesian thinking” to circumvent common cognitive pitfalls and exercise critical thinking more effectively. Librarians can definitely draw upon this book for examples to use in their information literacy classes.

Another relevant resource is this TED talk “Fake videos of real people – and how to spot them“, in which computer scientist Supasorn Suwajanakorn shows how he used AI and 3D modeling to create photorealistic fake videos of people synced to audio. It’s jaw dropping – those fake videos look incredibly authentic and I couldn’t tell the difference at all. This technology further complicates the information landscape what’s already littered with counter knowledge, false information and fake news. Maybe librarians can use this video to show their students what a dire situation we are in when it comes to online information, and critical thinking is thus indispensable in order to survive/thrive in such an information climate.

Time magazine published a thoughtful article on this topic as well – “A real fake news crisis“. This article equates the widespread misinformation online to “a public health crisis” and recommends the approaches that professional fact checkers use to sift through the internet, such as lateral reading and click restraint. When librarians devise their information literacy lesson plans, this article can be a valuable reference.

Recent trip to ACEID 2018 Conference

I attended the 2018 Asian Conference on Education and International Development (ACEID) last week, and it was my first time attending a conference outside of LIS. The majority of the conference participants (from over 40 countries) were from education, and their presentations covered a wide range of education related topics. My presentation was about how librarians and faculty collaborate in higher education, and it was scheduled in a session along with two other presentations focusing on higher ed. Unfortunately, not many people showed up in the audience – the concurrent sessions on K-12 issues outcompeted us. Still, I had a nice discussion with the small audience – before I presented my study, I showed them slides of librarian stereotypes and their true images (e.g. portraits of the book “This is What a Librarian Looks Like”, the New Zealand librarians cosplaying the Kardasians, and examples of the cool things that librarians made happen in the library), and the audience was quite impressed. They agreed that my presentation gave them good ideas of how to reach out to their librarians for help now that they know libraries are such an important component of the learning ecology in higher ed.

I had much fun interacting with education researchers at the conference and learned a lot myself. There’s a even research about how PMS (pre-menstrual syndrome) affects middle school girls’ academic performance in school. How interesting.

The conference took place in Kobe Japan during the beautiful cherry blossoms, and there were lines of cherry trees right outside of the conference venue, so I got to witness the fleeting and delicate beauty that has such significant cultural symbolism in Japan. It was indeed quite a sight!

A recent research/library trip to China

Last week I went to Shanxi Province, China for a research trip.

I’m collaborating with a Chinese professor from Shanxi University to study the role of libraries in helping the public fulfill their consumer health information needs. On this research trip, we conducted focus groups and in-depth interviews to gather data on people’s consumer health information seeking behavior and librarians’ preparedness in providing consumer health information service. It was a fruitful trip.

I visited three libraries – I wanted to post the photos here but couldn’t (maybe there were too many; hmm…maybe I should switch to a different blog platform?). So I ended up posting them on a separate webpage. We have students taking the international librarianship course in our program and maybe they will enjoy my post. 🙂