AI literacy exercises
Exercises: activities that teach students to question generative AI critically, in three learning lines of rising level.
These exercises come from the AI-Literacy Companion, Teacher Edition by the VU Centre for Teaching & Learning, under a Creative Commons Attribution licence. They are arranged in three learning lines that rise from understanding how AI works, through academic practice, to disciplinary judgement.
The companion builds on the open educational resources of Pennsylvania State University. The Dutch editions are our own translations; where they differ, the English text is the source.
An exercise is a kind of didactic tip too. Shorter advice on teaching with generative AI sits with the didactic tips on generative AI.
Further reading
Three books on assessment and teaching with AI. They are not part of this collection — they are whole publications with their own licence — but they sit well beside these exercises.
- Kevin Yee, Kirby Whittington, Erin Doggette & Laurie Uttich (2023). ChatGPT Assignments to Use in Your Classroom Today. FCTL Press, University of Central Florida. Over sixty ideas for prompting, searching, writing and studying with AI. CC BY-NC-SA 4.0.
- Kevin Yee et al. (2026). Show Your Work: Assessment in the Age of AI. FCTL Press, University of Central Florida. Sixty-two assessment ideas in three parts: co-creating with AI, forms AI cannot easily take over, and designing assignments to resist it. CC BY-NC-SA 4.0.
- Annette Vee, Marc Watkins & Derek Bruff (2026). The Norton Guide to AI-Aware Teaching. W. W. Norton. Fourteen chapters in two parts: planning an AI-aware course, and teaching one — from goals and policy through to assignments, activities, assessment and academic integrity. Pointedly not prescriptive: it helps you choose between embracing AI, setting firm boundaries, or something in between. No open licence; free as an ebook for instructors who use Norton textbooks.
The second book informed the AI resilience facet on the assignment and assessment formats.
Academic Integrity Exploration
Develop students' understanding of academic integrity principles in the context of AI technologies and establish clear ethical frameworks for responsible use.
Agree the ground rules for AI in your course
Settle explicitly what is and is not allowed in your own course, so that expectations are shared before the first assignment rather than argued about after it.
AI-Generated Hypothesis and Experimental Design Evaluation
Develop students' ability to critically assess research design proposals, identifying methodological flaws and strengthening scientific rigor.
AI-Generated Protocol Assessment
Train students to identify practical, technical, and scientific flaws in experimental protocols through systematic evaluation.
AI-Generated Summary Check
Develop students' ability to critically evaluate AI-generated content against primary sources through systematic comparison.
Ask students about their AI experiences
Find out what students already do with AI, so the course builds on their actual use rather than on what you assume it is.
Assessment of AI-Based Data Analysis
Develop rigorous critical evaluation skills for AI-generated data interpretation, statistical analysis, and scientific reasoning.
Cognitive Offloading Discussion
Examine the cognitive, educational, and societal implications of delegating mental tasks to AI technologies.
Comparison with Authoritative Sources
Train students to verify AI-generated information against established scholarly sources using rigorous comparison methods.
Critical Evaluation of AI Experimental Predictions
Develop students' ability to assess the scientific reasoning, mechanistic logic, and plausibility underlying experimental predictions.
Cultural Communication Bias
Reveal cultural and demographic biases in AI-generated advice and recommendations.
Data Privacy and AI Infrastructure
Increase awareness of data flows, storage practices, and privacy implications when using AI systems.
Discipline-Specific Diagram Evaluation
Develop domain-specific critical evaluation skills through systematic analysis of AI-generated visual and conceptual representations.
Do something impossible with AI
Students take on a persuasive task they would not trust themselves to pull off, plan how AI might help, document what actually happened, and report where the tool fell short.
Evaluation of AI-Generated Lab Notebook Entries
Assess AI's capability to accurately and completely document scientific work, and understand the limitations and risks of AI-assisted documentation.
Exercise: AI scavenger hunt
Students develop practical skills in identifying stylistic and substantive markers of AI-generated text, practise source evaluation in a low-stakes collaborative setting, and reflect on why AI-generated text can appear credible even when it…
Extended Topics for Advanced Discussion
The following topics can be developed into structured exercises using the pedagogical formats described above:
Feedback and Grading Analysis
Examine the reliability, consistency, and pedagogical value of AI-generated feedback and grading.
Five critical questions before using AI
Give students five questions to ask themselves before they open an AI tool, so that critical AI literacy becomes a habit of judgement rather than a technical skill.
Generative AI contested statements line-up
Students articulate and defend an initial position on a contested statement about AI, become aware of the diversity of views within the group, and practise the habit of demanding evidence for their own intuitive positions.
Learning Styles Myth Investigation
Demonstrate how AI systems can perpetuate educational myths and develop students' capacity to identify and challenge misinformation. Educational myths are an example of the process that old and outdated research and research findings remain…
PAIRR: peer and AI review + reflection
Put peer feedback before AI feedback on the same draft, then have students weigh the two against each other, so that AI review becomes something they judge rather than something they collect.
Peer Review of AI-Generated Research Proposals
Develop scientific writing skills and critical evaluation capabilities through systematic comparative analysis of AI-generated and human-authored research proposals.
Performance Prediction Bias
Explore how AI systems encode and reproduce gender and cultural biases in predictive outputs.
Prompt Engineering Challenge
Demonstrate the difficulty of achieving reproducible AI outputs, highlighting the inherent variability in AI systems.
Reading the titles aloud
Have students read out the titles of the drafts they just handed in. When the AI-assisted ones turn out to be near-identical, the point about generic thinking makes itself.
Summary Variation Analysis
Reveal the variability and inconsistency in AI-generated outputs, helping students understand that AI systems do not produce uniform or necessarily reliable results.
The analog sandwich
Structure a whole course as three units — AI-heavy, fully unplugged, then student choice — so that students can feel for themselves what each way of working does to their thinking.
The e-bike metaphor: relieving, strengthening or shortcutting
Use the e-bike as a shared image for what AI does to effort, so the discussion moves from whether AI is good or bad to which thinking you want to keep doing yourself.
The rubber duck and the AI student
Practise three ways of talking to a chatbot that support learning, against the one that does not, so students stop treating a fluent answer as evidence that they understand.
The transparency statement
Attach a three-line declaration of AI use to one assignment. The smallest possible move, and the only one that brings the subject into the open without surveillance.
Uncovering Visual Bias
Increase students' awareness of representational biases embedded in AI-generated images.
What uses more: the footprint of a prompt
Students compare the energy and water their own AI use consumed against everyday reference points, using a public calculator, and then read one critical text alongside the numbers.