Vrije Universiteit Brussel

04/02/2026 | News release | Distributed by Public on 04/03/2026 01:22

“Clever with AI?!”

The VUB Children's University is launching "Clever with AI?!", an online teaching package for pupils in Years 5 and 6. It allows children to explore how artificial intelligence works, the opportunities the technology offers, and where a critical approach remains essential. The package was developed in collaboration with Marjon Blondeel, a researcher with the VUB's Artificial Intelligence research group.

The full "Clever with AI?!" course is available through the VUB Children's University.

What should children take away most from "Clever with AI?!"?
Marjon Blondeel: "That AI is not some kind of magic box. It's built by people, usually works on the basis of data, and can therefore make mistakes. Many children are already using AI without realising it: the spam filter in their email, facial recognition on their phone, a robot vacuum that learns its way around the living room. AI systems also help doctors in hospitals to interpret medical scans. We want them to understand that AI may seem clever, but it doesn't 'understand' in the way humans do. It recognises patterns. That's why we teach children to ask critical questions: is this answer correct, is this image real, why is the algorithm showing me this video?We also highlight the environmental impact. Large AI models run on data centres that consume vast amounts of energy and water. Using AI responsibly also means: only use it when it genuinely serves a purpose."

How does the teaching package tie in with the research you carry out at the VUB AI Lab?
Marjon: "Our lab has worked in an interdisciplinary way for many years. Computer scientists, engineers, psychologists and philosophers collaborate to study how AI works and what its societal implications are. In 2020, we helped establish the AI Experience Centre, a demonstration and testing space where businesses, researchers, policymakers and the general public can experience AI first-hand. We showcase research demos there, as well as educational applications that explain how algorithms learn and where the risks lie.

"We compare how a robot vacuum learns with how children learn to ride a bike: trial, error, adjustment"

We also provide AI education at the VUB, from undergraduate and postgraduate modules to the Bachelor's degree in AI. In addition, we offer AI training for professionals and workshops for teachers on generative AI. 'Clever with AI?!' fits perfectly within that educational strand: making scientific knowledge accessible to everyone."

How do you make complex AI understandable for 9- to 12-year-olds?
Marjon: "By connecting it to experiences they already recognise. We compare how a robot vacuum learns with how children learn to ride a bike: trying, falling, and adjusting. We explain a spam filter by likening it to toddlers learning to recognise colours, purely through examples.

We also use activities that don't involve a computer. Think of a physical decision tree: by asking yes-or-no questions, you try to guess as quickly as possible which animal someone has in mind. In this way, children experience how algorithms draw conclusions. We also deliberately let AI make mistakes. If an app fails to recognise a straight banana as a banana, it creates a powerful learning moment: AI doesn't think, it follows patterns in its training data. Children grasp that surprisingly quickly."

Marjon Blondeel

"Young people are increasingly using chatbots as conversational partners or confidants"

Which ethical questions do you discuss with this age group?
Marjon: "Honesty is central. Many children think computers are neutral, but AI simply reproduces human biases. If an algorithm only 'sees' male doctors, it will conclude that doctors are men. That's the kind of bias they need to learn to recognise. We also talk about authenticity. In a world of deepfakes, cloned voices and AI-generated images, the question 'can you believe everything you see or hear?' becomes essential. We also discuss autonomy and freedom. What does it mean to be free when we increasingly follow AI-generated advice? Many of today's popular AI tools are designed primarily to serve corporate interests. Children need to understand that technology is not neutral, and that human values must always come first."

Do you see differences in how young people and adults use AI?
Marjon: "Yes. Adults tend to use AI in a more functional way: to rewrite texts or gather information more quickly. Young people, on the other hand, are more likely to use chatbots as conversational partners or even as confidants. Chatbots such as ChatGPT are designed to tell users what they want to hear. They give you the impression that you're always right. This can undermine a user's autonomy and their ability to think critically. Systems that aim to please therefore risk reinforcing bias and spreading misinformation.There is also an important learning process at stake. Adults developed their basic skills without AI. Young people are now developing those skills in a world where AI is embedded everywhere. If they outsource too many tasks too early, they miss out on the essential thinking processes that enable deep learning."

AI made accessible

With "Clever with AI?!", pupils aged 9 to 12 are introduced to the basic principles of artificial intelligence. They train a simple AI model, learn how to formulate clear prompts, and reflect on AI ethics. In doing so, they develop both digital skills and critical thinking. The teaching package was developed in collaboration with Marjon Blondeel, a researcher in the Artificial Intelligence research group at the Vrije Universiteit Brussel.

Vrije Universiteit Brussel published this content on April 02, 2026, and is solely responsible for the information contained herein. Distributed via Public Technologies (PUBT), unedited and unaltered, on April 03, 2026 at 07:22 UTC. If you believe the information included in the content is inaccurate or outdated and requires editing or removal, please contact us at [email protected]