What the 2026 ALA AI Literacy Research Means for MLIS Students

At the 2026 ALA Annual Conference in Chicago, University of Wisconsin-Milwaukee School of Information Studies assistant professor Wan-Chen Lee and doctoral candidate Juliana Hirt presented a poster, "Cultivating AI Literacy in Libraries," a resource guide for librarians and users.

AI literacy here means judging outputs for bias, privacy, and transparency, not just using a tool. For MLIS students and patrons, that changes what entry-level training and library workshops should cover.

The guide turns those principles into practical daily steps without demanding a single AI vendor or policy. By 2026, AI literacy is an expected professional competency in the library job market 2026.

What Is AI Literacy in Academic Libraries?

There is a real difference between learning to operate an AI tool and learning to judge whether that tool belongs in a library. In academic libraries, AI literacy usually means the second: a critical, values-based skill, not a technical certification.

A Library Definition, Not a Computer Science One

ACRL, the Association of College and Research Libraries and a division of the American Library Association (ALA), defines AI literacy in its AI Competencies for Academic Library Workers as the ability to "understand, use, and think critically about AI technologies and their impact on society, ethics, and everyday life."¹ Approved in October 2025, the framework does not require librarians to adopt AI for every purpose.¹ It treats AI as something to evaluate against library values.

Where Academic Libraries Fit

Academic libraries already host information literacy instruction, research consultations, and conversations about authority and bias. Adding AI literacy fits that mission. The ACRL framework organizes competencies into four areas: Ethical Considerations, Knowledge & Understanding, Analysis & Evaluation, and Use & Application.¹ These categories help library workers judge AI tools and other technology in libraries in context, from citation support to privacy risks.

How It Relates to Other Literacies

Digital literacy focuses on using devices, software, and platforms. Information literacy, rooted in ACRL's 2015 Framework for Information Literacy for Higher Education, focuses on evaluating sources and authority.² AI literacy extends both: it asks whose interests shape an AI model, what data trained it, and whether a generated answer is appropriate for a library user or assignment. In practice, academic librarians often teach all three together, but AI literacy adds attention to transparency, algorithmic bias, and the ethics of AI in libraries, especially the societal consequences of automated decisions.

Key Findings From the 2026 ALA Research Poster

The 2026 ALA Annual Conference poster by Wan-Chen Lee and Juliana Hirt makes one point unmistakable: AI literacy is now a core responsibility for libraries, not a fringe experiment. Presented June 25 to 29, 2026 in Chicago and published on July 15, 2026, "Cultivating AI Literacy in Libraries: A Resource Guide for Librarians and Users" lays out a practical path for the profession.

Opportunities Meet Challenges

Lee, an Assistant Professor at the University of Wisconsin-Milwaukee School of Information Studies, and Hirt, a SOIS doctoral candidate, framed AI as both a research and service opportunity. Their work highlights how AI can support library research and everyday library services, while naming the risks that come with it: bias in training data, privacy concerns around user data, and a lack of transparency in how AI systems reach conclusions. The poster treated AI not as a cure-all or a threat, but as a professional skill that must be taught, practiced, and updated.

A Guide Built for Everyday Decisions

At the center of the poster is a curated resource guide. The guide is designed to help librarians evaluate and apply ethical AI practices in daily work, from selecting tools to explaining AI to patrons. It is not a theoretical manifesto. It supports the kind of informed decision-making that libraries need as AI tools become more common in reference, instruction, and collections work.

For MLIS students charting MLIS career pathways, the takeaway is direct. The resource guide offers a model for how to translate AI literacy research into frontline library service and early career tips for librarians. Read the UWM SOIS announcement for the full details.

How MLIS Programs Are Teaching AI Literacy

The University of Arizona's LIS 528, Applied AI Literacies in Information Practice, is the clearest explicit AI-focused MLIS course in 2025-2026 catalogs. The course covers AI in scholarly research, automated content analysis, prompt engineering, teaching with AI, and ethical and equity issues. Supporting courses such as LIS 518, LIS 520, LIS 529, LIS 547, LIS 578, LIS 584, LIS 587, and LIS 671 reinforce adjacent skills, but the catalog does not state whether these are required or elective.

Explicit courses and supportive electives

At the University of Washington's MLIS program, courses like LIS 511, LIS 572, LIS 568, and LIS 591 appear in Spring 2026 schedules, but none are labeled as AI literacy. Emerging courses may exist, but the 2025-2026 results do not name them explicitly. This gap means many MLIS students still encounter AI literacy through professional development rather than degree coursework.

Professional development filling the gap

ALA and ACRL offer structured learning outside the degree path. ACRL's "Exploring AI with Critical Information Literacy" ran May 18-June 12, 2026, and ALA offers a free 8-week Generative AI Literacy Course. These programs emphasize critical AI literacy, algorithmic bias, privacy, and teaching patrons to evaluate AI outputs, which align with emerging librarian competencies in prompting, workflow use, and fairness.

UWM SOIS as a research-to-practice model

At the 2026 ALA Annual Conference, UWM SOIS Assistant Professor Wan-Chen Lee and doctoral candidate Juliana Hirt presented a curated resource guide for ethical AI use in libraries. The guide turns research into practical support for librarians, showing MLIS students how faculty-doctoral collaboration can produce tools that directly shape library services.

What assignments look like across the field

Catalog descriptions do not list specific LIS 528 assignments, but common patterns emerging from library AI education include bias audits, prompt engineering exercises, AI policy analysis, and building resource guides. The University of Virginia Library's Fall 2026 seminars, while not MLIS coursework, model project-based learning with critical evaluation and limits, reinforcing the competencies ALA and ACRL are beginning to articulate.

How to Measure AI Literacy After Library Instruction

Assessment after AI literacy instruction should focus on demonstrated judgment, not self-reported comfort. A measurable outcome is a patron checking whether an AI-generated citation actually exists, or explaining why an AI summary misstates a source, a task closely related to evaluating primary sources for information literacy. Recent reviews caution that questionnaires and surveys remain the most common method², and a 2026 analysis found 93.9% of AI literacy studies relied on subjective self-report instruments¹, which can overstate competence. Task-based performance is a better fit for library technology career work, where learners have to evaluate and revise AI output in context.

Choose Outcomes That Require Judgment

Six repeatable outcomes appear across 2024-2026 literature: recognize AI in a research workflow, distinguish AI output from source-based evidence, evaluate accuracy, bias, and limitations, apply privacy and ethical guidelines, revise or reject AI output to match a research need, and explain at least one risk such as hallucination, bias, or data exposure³. Avoid assessments that only ask whether patrons feel confident.

Build a Short Mixed-Method Package

A practical package can include a 5-to-10-item quiz, one scenario that asks learners to evaluate an AI-generated summary, one rubric-scored revision or critique, and one reflection prompt on ethics, bias, or privacy². A pre/post comparison is optional but helps show change. Document scoring rules and check interrater consistency, since reliability reporting is often inconsistent².

Keep Rubric Criteria Simple

A four-level rubric moves from Beginning to Advanced⁴. Two useful criteria are "Accuracy and evidence use" and "Bias and limitation detection." For accuracy, Beginning might recognize AI output without checking sources, Proficient identifies a specific missing citation or invented fact, and Advanced revises output to align with the source. For bias and limitation, Proficient names a specific limitation and its implication, while Beginning only states that AI can be biased.

Related Articles

How Public, Academic, and School Libraries Tailor AI Literacy

AI literacy is not one-size-fits-all. Public, academic, and school libraries teach the same core skills but shape them around the people in front of them. The difference is not whether to teach AI, but which questions matter most in each setting.

Public Libraries: Everyday Utility

Public libraries focus on broad, low-barrier learning. Programs in 2024-2026 include one-time "AI 101" classes, pop-up demonstrations, and task-based workshops on job searching and résumé writing. Mesa Public Library offers a free collaborative training program for residents, while the Urban Libraries Council has supported pop-up AI programs and formal introductory classes. The Public Library Association curates AI policies, staff training, and patron training examples. A March 24, 2026 ALSC webinar, "Future-Ready Youth: Preparing Our Patrons to Navigate the Age of AI," shows the field expanding toward youth. The guiding question: What is AI, where do you encounter it, and how can it help with everyday tasks?

Academic Libraries: Research-Centered Evaluation

Academic libraries take a methodical approach tied to scholarship. Guidance from ALA and ARL emphasizes selecting tools designed for accessibility and usability, citing AI outputs, applying bias evaluation frameworks, and building prompt toolkits. Choice360's competencies include "Should We Use AI?," "Prompting as Process," and "Verification as Scholarly Responsibility." The focus is on evaluating tools, documenting use, and applying AI responsibly in research.

School Libraries: Developmental Guardrails

School librarians adapt AI literacy into age-appropriate, scaffolded instruction. They emphasize verification, appropriate use, and source tracing, often through activities like a "Should we use AI?" decision exercise. Programs include University of Oklahoma's maker-based generative AI for kids in public libraries and University of South Carolina's open-access high school curriculum. Because students are minors, school librarians also add privacy and child-safety safeguards, teaching learners when AI is appropriate for a class assignment and how to verify its output.