I advocate for AI in education loudly, but I keep saying that advocacy without pedagogy is just installing software. So when a paper comes along that takes the conceptual foundations of AI literacy seriously, I read it carefully. Chiu’s (2025) editorial in Interactive Learning Environments is exactly that kind of paper: short, opinionated, and trying to clean up some terminology that has been getting muddled in the field since generative AI took off.
His core move is to separate two terms that are often used as if they meant the same thing: AI literacy and AI competency. According to Chiu, they operate at different levels of understanding and skill, and treating them interchangeably is causing real problems for both research and classroom practice.
AI Literacy and AI Competency Are Not the Same Thing
Chiu defines AI literacy as a foundational understanding of AI: what AI systems are, how they work, where they show up in daily life, and what societal questions they raise around bias, privacy, and economic disruption. Literacy doesn’t require coding skill. It requires curiosity and the willingness to engage critically with what AI does and what it can’t.
Competency is the next layer. It moves from understanding to doing. A competent person can use AI tools to accomplish specific tasks ethically, productively, and with self-reflection about what’s working. Competency is contextual. It depends on the field someone works in.
Chiu’s framing of the relationship is sharp. He writes: “While literacy asks, ‘What does this AI do?’ competency asks, ‘How can I make this AI work better?'” (p. 3226). And this might be my favorite line in the whole editorial: “Literacy is the compass; competency is the engine” (p. 3226). One word change reshapes the whole argument. You cannot optimize a system you do not understand. So competency rests on literacy.
The Ten Literacies That Actually Build AI Literacy
The most ambitious move in the editorial is the argument that AI literacy is fundamentally interdisciplinary. Chiu identifies ten supporting literacies he sees as the foundation for AI literacy: mathematical, data, ethical, media, computational, linguistic, visual, domain-specific, scientific, and design literacy.
This part of the paper deserves attention because it explains why a lot of AI literacy efforts feel disconnected from how AI actually shows up in classrooms. We treat AI literacy as a self-contained skill. Chiu argues it’s the opposite. Without data literacy, students cannot judge dataset quality. Media literacy is what lets them spot AI-generated misinformation. Ethical literacy is the difference between thoughtful use and a moral vacuum.
The interaction across literacies is what makes this framework useful. Bias detection pulls on data analysis, mathematical reasoning, and ethical thinking simultaneously. Prompt engineering pulls on language, computation, and the model’s training. Chiu warns that “[g]aps in any of these foundational literacies create vulnerabilities – misunderstanding capabilities, misinterpreting outputs, overlooking biases, or failing to grasp societal consequences” (p. 3227).
This connects to Hillman, Holmes, and Duarte’s (2025) rapid review of AI literacy frameworks for the Royal Society, which found that most existing frameworks treat AI literacy as if it were a self-contained skill, separable from other literacies. Chiu’s editorial is a useful corrective.
How Competency Develops
Chiu describes the development pathway from literacy to competency as a sequence: literacy first, then deliberate practice, then critical refinement of how AI is being used, then ethical integration, then competency. He is clear that competency does not emerge from simply using tools more often. It emerges when use becomes intentional and reflective. This tracks with what other recent work has been showing. Sidra and Mason (2026) make a similar argument in their work on collaborative AI literacy: skill develops when students reflect on AI outputs, not when they accept them.
What I Agree With and What I’d Push On
I agree with how Chiu separates literacy from competency. It maps onto what teachers I work with experience daily. A student who can name three AI tools is not the same as a student who can use one of them to support their learning ethically. Schools can use this split to build curriculum that doesn’t stop at “students should know about AI.”
I also agree with the interdisciplinary argument. Chee, Ahn, and Lee’s (2025) competency framework comes at a similar conclusion from a different angle: AI competency cannot be developed in isolation from the disciplines students are actually learning.
Where I’d push is on the practical translation. Chiu lists ten foundational literacies and says gaps in any one create vulnerabilities. That’s a fair point. But a classroom teacher with twenty-five students and forty-five minutes does not have the time or training to assess all ten.
The editorial stays at the conceptual level. It does not show what literacy-to-competency development looks like on a Tuesday morning. That gap is real, and it is one of the reasons I built the AI Use Agreement templates for elementary, middle and high, and higher ed contexts. Frameworks need scaffolding teachers can actually use.
The five future research directions Chiu identifies, especially equity and access in the literacy-to-competency pipeline, are the right questions. The hard work comes next. We have to move from definitions into curriculum, teacher training, and assessment design that helps teachers in the room. Compass without engine goes nowhere. Engine without compass goes off the road. Both have to be built, on purpose.
References
- Chee, H., Ahn, S., & Lee, J. (2025). A competency framework for AI literacy: Variations by different learner groups and an implied learning pathway. British Journal of Educational Technology, 56, 2146-2182. https://doi.org/10.1111/bjet.13556
- Chiu, T. K. F. (2025). AI literacy and competency: Definitions, frameworks, development and future research directions. Interactive Learning Environments, 33(5), 3225-3229.
- Hillman, V., Holmes, W., & Duarte, T. (2025). A rapid review of AI literacy frameworks, with policy recommendations. A report prepared for the Royal Society. London: The Royal Society. https://royalsociety.org/-/media/policy/projects/ai-in-education/hillman-et-al-a-rapid-review-of-ai-literacy-frameworks.pdf