When most schools hear "AI literacy," they look at the Computing department. This makes sense. AI is technology. Computing teaches technology. The Computing department should handle it.
The problem is that AI literacy is not a technology skill. It is a thinking skill. And thinking skills do not belong to one department.
A student who can explain how a neural network works but cannot evaluate whether an AI-generated news article is trustworthy has technical knowledge without literacy. A student who has never written a line of code but can spot when an AI is being confidently wrong has more AI literacy than the first student in the way that matters most.
AI literacy belongs across the curriculum for the same reason literacy itself does. Every subject teaches students to read and write. No school would say "reading is the English department's responsibility." AI literacy works the same way. Every subject has something to teach students about AI, and every subject teaches it differently.
What cross-curricular AI literacy actually looks like
In History, AI literacy is about evaluating AI-generated sources. Can the AI fabricate a convincing historical quotation? Yes. Can a student who has studied source evaluation spot the fabrication? Sometimes. That gap is where the learning happens.
In Art, AI literacy is about authorship and creativity. If an AI generates an image based on a prompt, who is the artist? What does originality mean when a machine can produce ten variations in seconds? These are not Computing questions. They are Art questions.
In Science, AI literacy is about data and bias. If a medical AI was trained mostly on data from one demographic, what happens when it diagnoses a patient from a different group? This is not abstract. It is happening right now. A Science teacher can use this as a case study without needing to explain how the AI model was built.
In Business Studies, AI literacy is about delegation and accountability. Which business decisions should be given to AI? Which should stay with humans? What happens when an AI makes a hiring recommendation that turns out to be biased? The Business Studies teacher is better positioned to teach this than the Computing teacher because the context is commercial, not technical.
In PE, AI literacy surfaces through performance analysis. AI systems now analyse match footage, track player movement, and predict injury risk. If the training data comes predominantly from men's football, how accurately does the system analyse a women's match? This takes thirty seconds to set up as a discussion question and connects to concepts students already know.
In Music, AI literacy raises questions about composition and copyright. If an AI generates a melody by learning from thousands of existing songs, is the output original? What if it sounds similar to a specific existing song? These questions are already part of Music education. AI just adds a new angle.
None of these examples require a laptop. None require the teacher to understand how AI is built. All require the teacher to ask a good question about AI in the context of their subject.
Why the Computing department cannot do this alone
The Computing curriculum teaches students how AI systems work. Algorithms, training data, machine learning models, neural networks. This is essential knowledge. It is also insufficient.
Knowing how a car engine works does not make you a good driver. Knowing how a neural network works does not make you a critical user of AI. The critical use happens in context, and the context is the subject the student is studying.
A Computing teacher can explain that AI systems learn from training data. A History teacher can show what happens when the training data reflects the biases of the people who collected it. A Geography teacher can show how those biases affect which communities appear in mapping data. An English teacher can show how those biases affect the language AI uses to describe different groups.
Each teacher adds a layer that the Computing teacher alone cannot. Not because the Computing teacher lacks knowledge, but because the subject context is where the AI literacy concept becomes real to the student.
The OECD AILit Framework, which will shape the PISA 2029 assessment, says it directly: educators should embed AI literacy when and where it aligns with their subject and context. The DOL framework says the same: learning should be embedded in context. UNESCO's scenarios span subjects. Every major framework agrees: AI literacy is cross-curricular.
How to start without rewriting your curriculum
You do not need a new scheme of work. You do not need a department initiative. You do not need to coordinate with anyone. You need ten minutes and one good question.
Pick a lesson you are teaching this week. Ask yourself: is there a moment where I could ask students to think about AI? Not use AI. Think about it.
If you teach a lesson involving data, you can ask: what would happen if an AI made this decision instead of a human? If you teach a lesson involving evaluation, you can ask: could an AI produce work that looks like this, and how would you know the difference? If you teach a lesson involving design, you can ask: if you had to write instructions for an AI to do this task, what would go wrong?
That question, asked with your subject knowledge and your understanding of your students, is an AI literacy activity. It takes no preparation. It requires no technology. It fits inside what you are already doing.
If you want more than one question, AILitKit generates a full set of activities from any lesson in your scheme of work. A Lesson guide takes your actual lesson content, finds the AI literacy connections, and gives you activities with a script for introducing the concept, a question students will actually ask (with a suggested response), timing, resources, and a contingency plan for when it does not land. Every activity maps to the major international frameworks automatically.
For heads of department who want to see how AI literacy maps across a full unit, a Topic guide analyses an entire scheme of work and shows which lessons have natural AI literacy connections and which ones honestly do not. It includes a department meeting agenda so you can brief your team in 15 minutes.
For school leaders who want the whole-school picture, a Whole Curriculum guide audits provision across subjects and key stages, identifies quick wins, and produces a governor briefing with framework coverage and next steps.
The permission you might need
Some teachers are waiting for permission. Permission from their head of department, their SLT, their exam board. They want to know it is safe to spend ten minutes on something that is not in the syllabus.
AI literacy is in the direction of travel for every major curriculum body. PISA will assess it in 2029. The OECD framework defines it. The UK curriculum review will address it. The UAE has already mandated it. The direction is clear. The only question is when, not whether.
If you are waiting for a formal directive, you may be waiting a while. If you try one activity this week, you will have more experience than 80% of teachers in the country. That is not a criticism of the 80%. It is a description of how early the window is.
The teachers who start now are not ahead of the curve. They are at the front of a very long queue.
AILitKit works for any subject, any key stage, any scheme of work. Give it a lesson. Get a guide with AI literacy activities built for your subject and your students. Try it free at ailitkit.com