Learning activity ideas for using AI with learners

The Creative classrooms with AI: Innovate responsibly course first ran on EU Academy in March 2026 and is still available in a self-paced format.

Participants explored how to use AI as a creative partner in the classroom, moving beyond content generation to support pupil-led creation, innovation and critical thinking. Through hands-on activities, they designed classroom-ready tasks using AI tools for creative expression across subjects and age groups, including prompt design challenges, science, technology, engineering, arts and mathematics (STEAM) projects, and interdisciplinary learning experiences that transform learners from passive consumers to active creators.

The course emphasised practical strategies to foster key competences such as problem-solving, collaboration and creativity, while embedding ethical and responsible AI use through intentional, transparent and reflective teaching practices that participants can immediately apply in their own educational settings.

The learning activities below offer valuable ideas and inspiration to teachers who want to use AI with their learners in a creative and responsible way. Following each activity, you will find additional ideas and approaches for adapting it to other classes.

The learning activities have been reviewed and curated by the course coordinator. A big thank you to the authors:

Ana Mihaela Pop, Giorgos Skoumpaflos, Helga Karljik, Kristina Vaičiulienė, Mariam Kapanadze, Slavica Bernatović

• Age group: primary, upper primary, lower and upper secondary education
• Subjects: all
• Published by: European School Education Platform
• Year: 2026
• Languages available: English 

AI-generated images as tools for textual analysis

Age category: 11-12 years

 Gustavo Fring / Pexels

A Romanian language and literature teacher adapted the ‘AI as Thinking Partner’ strategy for year 6 learners studying Ion Luca Caragiale's sketch ‘Mr Goe...’ by transforming AI image generation into a critical validation exercise. Instead of simply creating aesthetic character portraits, learners generated images of Goe at a train station platform and then acted as ‘literary detectives’, meticulously comparing AI outputs against textual evidence to identify interpretation errors such as anachronistic elements, foreign language inscriptions and historically inaccurate details.

This approach worked well because AI ‘hallucinations’ became powerful learning moments: pupils rigorously returned to the source text to argue their critical observations, demonstrating that technological errors can deepen textual understanding when guided by expert teaching. The strategy successfully developed digital literacy, critical thinking and close reading skills while reinforcing that AI serves as a creative starting point requiring human judgment and textual expertise to navigate its limitations.

What worked remarkably was the transformation of a technological ‘error’ (the AI ​​hallucination) into a moment of deep learning. I was surprised by the rigour with which the learners returned to the source text to argue their critical observations. This experience demonstrated that while AI can provide a creative starting point, the teacher’s professional judgment and expert knowledge remain essential to navigate through the technology’s unrealistic or vague suggestions.

— Ana Mihaela Pop, Author

Download the activity ‘AI-generated images as tools for textual analysis’.

Mastering prompt precision in historical writing

Age category: 11-13 years

Zen Chung/ Pexels

In this two-round prompt engineering challenge, 9th-grade history learners explore the liberation of a European city in 1944 by learning to transform vague AI outputs into precise, historically rich narratives.

Learners begin with a visual description of a liberated city square and first use a basic, imprecise prompt that generates a generic response. In the second round, they apply prompt engineering techniques by adopting the role of a war correspondent, specifying a newspaper report structure and requesting a formal tone appropriate for public journalism. Through this progression, learners discover how strategic prompt design – including contextual details like human solidarity moments (an elderly woman sharing bread with a soldier) – can guide AI to produce more vivid, emotionally resonant and historically meaningful content.

The activity culminates in a critical reflection where learners recognise that AI's apparent emotional depth stems from their own deliberate instructions, not the technology's independent understanding. This reinforces that they are active designers of knowledge rather than passive consumers of AI-generated content.

By forcing the AI to focus on human solidarity (e.g. the elderly woman gives a piece of bread to the soldier), leaners practise ethical judgment. They would realise that the AI’s output is seemingly emotional since we have given it a specific context, and not because of its own initiative. I expect learners to realise that AI does not replace their historical knowledge; rather, it enhances their ability to express that knowledge creatively. The learners become active designers and not passive consumers.

— Giorgos Skoumpaflos, Author

Download the activity ‘Mastering prompt precision in historical writing’.

A photography-based project of environmental futures in schools

Age category: 16-18 years

Rick Esquivel / Pexels

In this creative environmental project, learners aged 16-18 use photography and visual imagination to explore climate change through their immediate school surroundings. The activity begins with learners photographing familiar locations around their school, courtyards, gardens, playing fields or nearby natural spaces. They then engage in a ‘two-way reimagining’ task where they digitally recreate these same locations in two contrasting futures: a dystopian scenario showing environmental decay and neglect, and a utopian vision depicting ecological flourishing and sustainable transformation.

This playful experimentation allows learners to move beyond the paralysis of eco-anxiety by giving them creative agency to visualise both consequences and possibilities. Rather than producing traditional written reports, learners express their understanding of environmental cause and effect through visual storytelling, combining their passion for photography with deep analytical thinking about local and global environmental issues.

The project ends with a peer discussion where learners share their contrasting visions, explaining the environmental choices and actions that would lead to each outcome. This transforms abstract climate concepts into personally meaningful civic imagination rooted in their own lived experiences.

I recall a specific learner who was very talented at photography but struggled to engage with written reports. During a previous lesson, they took a beautiful photo of a local park but couldn't explain its environmental significance in a standard essay format. This moment reveals that traditional assessment often misses the creative potential of learners who think visually rather than verbally.

By linking this to the theory of ‘projects, passion, peers and play’, my school surrounding's allows this learner to use their passion for photography as a vehicle for complex environmental thought. The recreation phase of the project – where they imagine the landscape's decay or its flourishing – requires them to apply theoretical environmental knowledge in a way that feels personal and creative. This demonstrates the shift towards a learner-centred education where the learner's own vision drives the learning process.

— Helga Kraljik, Author

Download the activity ‘A photography-based project of environmental futures in schools’.

Prompt precision through creative AI design for young learners

 Age category: 9-11 years

RDNE Stock project / Pexels

In this hands-on prompt engineering activity, 5th-grade learners use Pixelbin AI to design Easter eggs for a colouring book as part of the eTwinning project ‘Eggventures’.

The activity begins with learners trying to generate designs without specific prompts, which typically produces simple, generic egg images. Through guided experimentation, learners discover that precise, detailed prompts dramatically improve their AI-generated designs. The teacher supports this learning process by providing example prompts and helping learners refine their descriptions, encouraging them to modify existing prompts or create entirely new ones. As learners add specific details about colours, patterns, themes or styles – sometimes changing just a single word – they witness immediate differences in the AI output, transforming basic eggs into creative, intricate designs suitable for their collaborative colouring book.

This repeated process of prompt adjustment, experimentation and refinement not only teaches learners fundamental digital literacy skills about how AI tools respond to human instructions, but also fosters creativity, critical thinking and engagement. Through playful trial and error, learners grasp a powerful lesson: that effective communication with AI requires precision and thoughtfulness, and that they hold the creative control to guide technology towards meaningful, personalised results rather than accepting generic outputs.

Working with prompts encouraged creativity and engagement.  Learners enjoyed experimenting with their own prompt adjustments and seeing better results.

— Kristina Vaičiulienė, Author

Download the activity ‘Prompt precision through creative AI design for young learners’.

Teaching ESL learners descriptive language precision through AI-powered cultural mashups

Age category: 7-13 years

CDC / Pexels

In this innovative English as a Second Language (ESL) activity designed for Georgian learners aged 7-13, pupils combine traditional Georgian cultural symbols with modern technological concepts to create imaginative ‘cultural innovation’ designs using AI image generation. Learners begin by selecting one traditional Georgian symbol – such as a Chokha (traditional Georgian clothing), a Kvevri (clay wine vessel) or other cultural artefacts – and pairing it with a contemporary technological function like space travel, digital gaming or robotics. The pedagogical purpose is twofold: to make abstract English vocabulary tangible and functional, and to transform grammar and descriptive language practice into a high-stakes creative challenge where linguistic precision directly determines visual outcomes.

Learners craft detailed English prompts describing their hybrid creations, using specific adjectives, nouns and descriptive phrases to communicate their vision to the AI image generator. When initial simple prompts like ‘fast Kvevri’ produce generic or unsatisfactory results, learners must refine their language – adding details about colour, texture, lighting, materials and specific functions – to achieve their desired visual outcome. This iterative process turns the AI into a ‘visual mirror’ that immediately reflects the accuracy and richness of their English descriptions, transforming what could be a dry vocabulary lesson into an engaging design activity where language becomes a powerful creative tool.

Through this cultural mashup approach, learners not only develop precise descriptive language skills but also explore the intersection of their heritage and contemporary innovation in a personally meaningful way.

I anticipate a significant boost in learner engagement. For younger learners (ages 7-11), abstract vocabulary often feels ‘dry’. However, when they see that saying ‘A golden Kvevri shaped like a rocket ship’ actually produces a picture, the language becomes functional and alive.

— Mariam Kapanadze, Author

Download the activity ‘Teaching ESL learners descriptive language precision through AI-powered cultural mashups’.

Visualising physics concepts with tinkering labs

Age category: 13-14 years

Cottonbro studio / Pexels

In this AI-enhanced physics tinkering lab, learners use AI illustration tools to visualise abstract physics concepts such as projectile motion, thermal expansion, velocity and trajectory by crafting detailed descriptive prompts.

The activity addresses a common challenge in physics education: making invisible forces and abstract principles tangible and understandable. Learners begin by selecting a physics concept they are curious about. They then write specific prompts that describe the physical scenario, such as a football being kicked or a rocket launching, and the scientific elements they want visualised, including vectors, forces, labels and visual style preferences. Working in pairs, learners generate images, compare unexpected results, provide feedback to one another and refine their prompts to achieve more accurate scientific representations.

This collaborative experimentation transforms AI from a passive answer-generator into an active creative partner that requires clear communication and critical thinking. When AI outputs don't match expectations, learners must analyse why, adjust their language and problem-solve together, deepening their conceptual understanding as they make their ideas visible. The activity positions learners as creators and critical thinkers who control their learning journey, building confidence in scientific vocabulary, visual interpretation skills and responsible AI use through hands-on exploration and reflection.

Learners were in control of their learning; they were creators, problem-solvers and critical thinkers. Using AI in a tinkering lab style encouraged curiosity, experimentation and reflection. I imagine that with repeated practice, learners would become more confident in communicating ideas clearly, understanding abstract concepts and collaborating to improve outcomes – all while using AI responsibly.

— Slavica Bernatović, Author

Download the activity ‘Visualising physics concepts with tinkering labs’.