Generative artificial intelligence is entering university classrooms rapidly, but much of the conversation still centres on cheating or assessment integrity. While these concerns matter, they risk overlooking another important educational question: how can students learn to engage critically with AI itself?
Many students are already experimenting with tools such as ChatGPT, Copilot and image generators. But using AI is not the same as understanding it. In my own teaching conversations with students, questions often arise around bias, authorship, originality and the ethical implications of generating AI content.
- Reaching for the pizza? How to help students make informed choices with GenAI
- ‘Our agency over our words is central to our agency over ourselves’
- Show students what thoughtful engagement with GenAI looks like
This raises a challenge for educators: how can we help students move beyond using AI simply as a productivity tool and instead begin questioning how these systems shape knowledge, creativity and representation?
One approach I experimented with was surprisingly simple: asking students to rewrite fairy tales with AI.
Why fairy tales?
Fairy tales are culturally familiar stories that carry strong assumptions about gender, morality and power. Because students already know these narratives, they provide an accessible starting point for analysing how stories are constructed.
At the same time, GenAI tools produce new narratives by recombining patterns from their training data. When students place these two narrative systems, traditional fairy tales and AI-generated outputs, side by side, it becomes easier to see how stereotypes, omissions and narrative conventions are reproduced.
In other words, fairy tales become a safe and familiar space where students can interrogate bias and authorship in both historical storytelling and contemporary AI systems.
The workshop
The activity took place as a hands-on interdisciplinary workshop with 11 students from different subjects and study levels.
Students used tools including Microsoft Copilot and Adobe Express to generate summaries, images and story continuations for familiar fairy tales. They then compared these outputs with human-written versions and their own interpretations.
The workshop included several simple tasks:
Compare human and AI summaries
Students asked AI to produce a short summary of a familiar fairy tale and compared it with a human-written version. This often revealed simplifications, missing details or shifts in emphasis. In one case, an AI-generated summary of Hansel and Gretel removed the breadcrumb trail entirely, simplifying the story into a much more generic narrative.
Generate AI illustrations
Students prompted AI to illustrate a scene from the story and compared the output with their own imagined version. This prompted discussion around representation, beauty standards and cultural assumptions. One student noted that AI-generated images tended to portray “good” characters as attractive and villains as visibly ugly, reinforcing familiar visual stereotypes within fairy-tale imagery.
Extend the narrative
Students asked AI to continue the story. Many noticed that the AI tended to produce predictable endings and moral lessons. One student observed that even when they tried to complicate the narrative, the AI pushed the story back towards a familiar redemption arc in which the villain felt remorse and changed their ways.
Rewrite the story
Finally, students revised the AI-generated text and images to challenge stereotypes or create alternative interpretations. Several students deliberately reworked AI outputs they felt were overly sanitised or morally simplistic, adding greater emotional complexity, ambiguity or more inclusive character portrayals.
One student reimagined Beauty and the Beast so that the Beast remained in his non-human form and the entire community transformed alongside him, embracing difference rather than returning to “normal”. Another rewrote Snow White so that the Prince and Evil Queen became allies driven by jealousy, while the dwarfs defended Snow White together. Others imagined Rapunzel opening a magical hair salon in 2025 or transformed Little Red Riding Hood into a skilled wolf hunter protecting vulnerable children.
Throughout the workshop, discussion and reflection were embedded into the process. The aim was not simply to learn how to prompt AI, but to question what the outputs revealed about how these systems work. Students were encouraged not only to identify bias or simplification, but to reshape narratives themselves, experimenting with alternative endings, unconventional characters and more inclusive forms of storytelling.
What students discovered
Before the workshop, most students reported low confidence using AI creatively. Many also expressed concerns about bias, originality and ethical responsibility.
After the session, students felt more comfortable experimenting with AI tools, but this increased confidence was accompanied by greater critical awareness.
Students quickly identified patterns in the AI-generated content. These included:
- stereotypical characters and gender roles
- Eurocentric representations and beauty standards
- simplified storylines and predictable moral endings.
Some students found it difficult to generate diverse characters. One student attempted to create a plus-sized wolf-hunting version of Little Red Riding Hood, but the image generator repeatedly produced slim, idealised characters instead.
These moments of frustration became some of the most valuable learning points. Rather than treating AI outputs as authoritative, students began asking why certain narratives kept appearing and whose perspectives might be missing.
By the end of the workshop, students often described AI as a useful but limited creative collaborator – helpful for generating ideas, but unable to replace human imagination or judgement. One participant compared AI to “a hammer”: a useful tool in the hands of a user, but incapable of generating meaning or creativity independently.
Why storytelling works for AI literacy
One reason the activity worked well is that storytelling makes abstract issues visible.
Concepts such as bias, representation and authorship can feel theoretical when discussed in the abstract. But when students encounter them directly in AI-generated narratives or images, they become much easier to recognise and debate.
The familiarity of fairy tales also reduces cognitive load. Because students already know the stories, they can focus on analysing how the narrative changes when AI becomes involved.
Perhaps most importantly, the workshop positioned AI as a shared object of enquiry, rather than a tool students must simply learn to use correctly. Students experimented, questioned outputs and discussed their observations with peers.
This collaborative approach helped students develop a more nuanced understanding of both the possibilities and limitations of generative AI.
Practical ways to try this in your own teaching
The activity can easily be adapted across disciplines and class sizes. A few simple strategies can help embed critical AI literacy into teaching:
- Start with familiar material: Using well-known texts, case studies or cultural narratives makes it easier for students to identify differences between human and AI-generated outputs.
- Encourage comparison: Ask students to compare AI-generated content with human-created work and analyse the differences.
- Embed reflection and discussion: Students develop deeper insight when they can share observations and debate ethical questions with peers.
- Focus on judgement, not just skills: Rather than teaching students how to use AI “correctly”, encourage them to consider when, why and whether AI should be used at all.
Moving beyond the AI panic
Generative AI is likely to continue shaping how students learn, create and communicate. Simply warning students about misuse or academic misconduct may not be enough to prepare them for this reality.
Instead, education has an opportunity to help students develop the confidence and critical awareness needed to navigate AI responsibly.
In this workshop, rewriting fairy tales helped students move beyond seeing AI as either a threat or a shortcut. Instead, they began to see it as something to question, challenge and shape.
Sometimes, the best way to start that conversation is with a familiar story.
India Lawton is a senior lecturer at Southampton Solent University.
If you’d like advice and insight from academics and university staff delivered direct to your inbox each week, sign up for the Campus newsletter.
comment