With the AI era in full swing, new tools and platforms are constantly appearing. This can make it difficult for educators to rely on fixed teaching materials or standardised outreach activities. For this reason, I believe that AI literacy should not simply be about teaching students how to use a particular tool. Instead, the most important goal is helping students to recognise that these technologies are accessible to them, and encouraging them to start experimenting, exploring and learning independently.
Students become far more engaged when they can directly interact with AI systems, create projects, experiment with ideas themselves and see results immediately rather than simply listening to explanations about how AI works.
Over the years, I’ve developed an AI literacy ecosystem that supports students progressively from primary school to university and beyond. Through outreach workshops, massive open online courses (Moocs), pre-university summer institute programmes, STEM outreach collaborations with local schools and university teaching, I’ve seen that AI literacy outreach works most effectively when students are introduced to technologies gradually across different stages of learning.
Across these different educational settings, four teaching principles have consistently guided my outreach activities:
- Try it before we explain
- What you see is what you get
- Prompt, improve and refine
- Spark, explore and continue
Together, these principles help students move from initial curiosity to independent and lifelong learning.
1) Introduce AI concepts through a ‘try it before we explain’ approach during early primary education
At the primary school level, outreach should focus less on technical depth and more on curiosity, confidence and accessibility. Young students do not need to fully understand machine learning models or programming syntax before engaging with AI-related technologies.
Instead, the goal should be to help students recognise that these technologies are already part of their daily lives and that they can interact with them creatively. In many outreach workshops, I introduce students to simple vibe coding platforms and the development of AI chatbot applications, using MIT App Inventor, which allows them to experience computational thinking through experimentation and play.
At this stage, I frequently adopt a “try it before we explain” approach. Rather than beginning with technical explanations of AI systems, I encourage students to experiment with simple tools and discover what they can create. After that, our discussions about the underlying ideas become much more meaningful.
Success is not measured by how much technical knowledge students remember. Instead, success comes from helping students to feel curious, confident and willing to continue exploring technology afterwards.
2) Build confidence and experimentation through a ‘what you see is what you get’ approach during secondary education
At the secondary school level, students are generally more ready to explore how AI and computing technologies can support problem solving, creativity and interdisciplinary learning.
I believe outreach activities at this stage should focus strongly on hands-on experimentation and visible outcomes.
- Turn AI from a magic box into a tool to be interrogated
- ‘If you like, I can…?’ Why GenAI needs to come with a health warning
- The key to designing outreach activities students remember
I have found that a “what you see is what you get” approach is particularly effective for AI outreach. Students become more motivated when they can immediately see the results of their work rather than spending long periods learning theory before creating anything themselves.
In many workshops, students build simple software applications with the help of AI, and they continuously test, refine and modify their applications through multiple iterations of prompting and experimentation.
3) Support interdisciplinary AI literacy through ‘prompt, improve and refine’ learning at university level
At university level, AI literacy outreach should become increasingly interdisciplinary. AI-assisted coding is no longer relevant only to computer science students. Students from business, engineering, science, social sciences and humanities are all beginning to use AI tools to support research, data analysis, content creation and project development.
For this reason, I have developed a Horizon AI Common Core course to encourage students to use AI-assisted coding to develop prototypes, analyse data, create digital artefacts and explore entrepreneurial or research ideas at the very beginning of their university journey.
Rather than focusing purely on programming theory, these activities encourage students to experiment with AI technologies in ways that are directly connected to their own interests, projects and future careers.
At this stage, students also benefit from what I call a “prompt, improve and refine” approach. Rather than expecting AI-generated solutions to be correct on the first attempt, students learn by crafting prompts, evaluating outputs, refining their requests and iteratively improving their work. Much of the learning occurs through this process of experimentation and refinement.
Instead of a separate thread to formal education, at this stage, AI literacy should become part of a broader ecosystem that connects outreach, university teaching, interdisciplinary projects and self-directed exploration.
4) Extend learning opportunities through a ‘spark, explore and continue’ approach
One major challenge in outreach is that short workshops alone are often insufficient for sustained learning. Students may become interested during an activity, but they still need opportunities to continue exploring afterwards.
I’ve noticed that students often continue exploring topics independently after their first exposure through outreach activities. Some students begin experimenting with their own projects, while others continue learning through online courses or additional workshops. In many cases, the initial outreach activity simply acts as a starting point that encourages students to explore further on their own.
For this reason, online learning resources and Moocs can play an important role in supporting long-term AI literacy development. Through my self-paced Moocs, students can continue experimenting with coding, AI applications and computational tools beyond classroom or outreach environments.
This reflects what I call a “spark, explore and continue” approach. Outreach activities should spark curiosity, provide opportunities for students to explore further and support them in continuing their learning journey independently.
Create an ecosystem rather than isolated outreach activities
One of the most important lessons I have learned is that effective AI literacy outreach should not rely on isolated events alone. Students benefit most when outreach activities are connected across different stages of education and are supported by opportunities for continued exploration.
Primary school outreach can introduce curiosity and accessibility; secondary school outreach can build experimentation and confidence; university education can support interdisciplinary application and deeper exploration; while Moocs and online learning platforms can provide opportunities for continued self-directed learning.
When these different components are connected, AI literacy outreach becomes a sustainable learning ecosystem rather than a series of disconnected activities.
Together, the four principles I’ve outlined here help students move from curiosity to experimentation, from experimentation to creation and, ultimately, from guided activities to independent lifelong learning.
Kenneth Wai-Ting Leung is associate professor of engineering education at the Hong Kong University of Science and Technology.
If you would like advice and insight from academics and university staff delivered direct to your inbox each week, sign up for the Campus newsletter.
comment