Beyond the Demo: How Library AI Instruction Can Shift from Literacy to Competency

Students know how to use AI. What they need to know is how it judge its use.

Posted on March 30, 2026 in Blog Posts

Over the past several years, academic libraries have responded to generative AI by developing AI literacy workshops that introduce core concepts, outline limitations, and demonstrate common tools. This work has been necessary and timely. As AI becomes embedded into academic workflows, a growing body of student feedback suggests that repeated orientation-style instruction can feel less supportive than intended. Students are often using AI tools for brainstorming, drafting, and clarification, and when instruction focuses primarily on demonstrations or misuse warnings, it can register for students as a lack of trust rather than guidance.

This tension is shaped in part by who attends library AI programming. Library workshops frequently draw faculty, staff, and researchers seeking clarity around policy and practice, while undergraduate students are often reached when librarians are invited into courses. Instruction framed around demonstration and risk management can undermine student engagement if learners experience it as policing rather than support. Students are “partners, not passengers” with AI and want to help shape how it is integrated into academics. Recognizing this shift is a starting point for reconsidering what the next phase of AI instruction should prioritize.

The Pedagogical Limits of Orientation-Style Workshops

Demonstrations and overviews remain effective instructional tools for introducing unfamiliar technologies. They lower barriers to entry, establish vocabulary, and help learners understand how a system works. In an earlier LibTech Insights article, I focused on this integrative challenge: how librarians could meaningfully incorporate GenAI into one-shot information literacy sessions without displacing foundational research skills. That work emphasized modeling AI as one step within the research process rather than as a replacement for scholarly inquiry.

But orientation-style workshops have pedagogical limits once students move beyond first exposure. Demonstrations reveal functionality, but they obscure the decision-making that governs academic use. When librarians select the tool, craft the prompt, and interpret the output, students observe the process without practicing it. This approach struggles to teach the judgment, accountability, or verification skills that are important in AI use. More sophisticated demos are not the answer to growing concerns about overreliance, misuse, or academic integrity. Research on AI in higher education notes that literacy-focused instruction must be complemented by opportunities for applied, reflective use if learning is to transfer beyond the classroom. AI instruction must shift from showing what AI can do to practicing how scholars decide when and how to use it.

---

🔥 Stay up-to-date with LibTech Insights by signing up for our free newsletter. Just one weekly email with our new blog posts, top tech news stories, and other bonus content. Check out some posts from our archive:

• Invisible Sources: Transparency and Attribution as Foundations of AI Literacy Instruction
• How to Find and Fix Hidden Accessibility Issues
• Creating Alt Text for Digital Collections with AI: A Case Study

💜 ALA is having its 150th anniversary. Come celebrate with us!

---

From AI Literacy to AI Competency

If AI literacy establishes awareness, AI competency emphasizes judgment. Instruction no longer focuses on whether students understand what generative AI is, but on whether they can make defensible academic decisions about when and how to use it. Competency‑oriented instruction focuses on practicing skills like weighing appropriateness, explaining reasoning, verifying outputs, and taking responsibility for use. These are not advanced technical skills. They are scholarly habits. Ethical reasoning and decision-making transfer whether students are using ChatGPT or a different tool. Positioning AI workshops around ethical competency allows librarians to support student learning while reducing the emphasis on constant technological change.

Libraries are well positioned to lead instruction in AI competencies: how to evaluate outputs, document reasoning, design responsible workflows, and understand verification as part of scholarship. Faculty and course instructors then define discipline-specific norms and assessment expectations. The strongest instructional model is partnership, with librarians teaching transferable scholarly processes, faculty defining discipline expectations, and students deciding when, how, and why responsible AI use is necessary.

AI Competency Activities

Reframing AI workshops around competency requires a shift in the instructional design. Rather than organizing sessions around tools or demonstrations, workshops are structured around moments where students make academic decisions and explain their reasoning. The following three activities illustrate how this shift can be enacted in short instructional contexts, whether in one-shot workshops or course-embedded sessions.

• The “Should We Use AI?” decision activity slows down default automation. Students are given a research or writing scenario and asked to decide whether AI use is appropriate at all. They justify their decision in relation to the task’s goals, expectations, and risks before selecting any tool or generating output. Learning occurs in the rationale and recognizing that nonuse can be an academic choice, not in reaching a correct answer. This activity usually doesn’t include using an AI tool.
  
  
• Prompting as Process demonstrates prompting as an iterative, reflective practice rather than a technical shortcut. Instead of presenting “effective prompts” as prepackaged solutions, students draft, revise, and compare prompts in relation to a specific academic task. The instructional emphasis is on how changes in framing, constraints, or specificity shape output and what those changes reveal about the underlying task.
  
  
• Verification as Scholarly Responsibility reframes checking AI output as a routine academic obligation rather than a response to AI. Students identify key claims in AI-generated content and trace them back to credible sources, documenting how they assessed accuracy, context, and relevance. Verification is presented as part of normal scholarly practice rather than as something only reserved for AI.

These activities shift the instruction from exposure to practice. They make student judgment visible and position AI use as an academic choice within their work. This isn’t limited to student instruction, but to faculty and staff in AI sessions where activities can prompt conversation about values and accountability within their different fields. This shift from demonstration to deliberation builds trust.

Moving Beyond AI Literacy toward Sustainable AI Instruction

AI literacy workshops are necessary and an important response to rapid technological change. However, as student use of generative AI is normalized, libraries are increasingly called upon to offer instruction that goes beyond demonstrating AI capabilities. A competency-based approach provides a way to do so without requiring additional tools, extended instructional time, or constant adaptation to new platforms. By centering judgment, verification, and accountability, competency workshops align library instruction with enduring academic values. They treat students as capable decision‑makers and positions AI use as an academic choice within disciplinary work.

Framing AI instruction around what students can do and how they can justify those choices positions libraries as partners in learning. As institutions continue to navigate AI’s role in academics, competency-based instruction offers a sustainable path forward that supports student agency, reinforces scholarly responsibility, and allows library teaching to evolve alongside AI.

Related Posts