Why AI Literacy Belongs Alongside Digital Literacy

Quick answer

AI literacy belongs alongside digital literacy because it is the natural next layer on the same foundation, not a separate subject competing for space. Digital literacy already taught students the why and when of technology - not merely the how - with a strong focus on privacy, security and online safety. AI literacy adds the specific judgement that generative tools demand: understanding why AI produces what it does, directing it well, evaluating its output, and using it responsibly. The relationship is sequential and complementary. Digital literacy governs a student's relationship with technology in general; AI literacy governs their relationship with a technology that generates content and can be fluently, confidently wrong. For Australian schools, framing AI literacy this way is the school-safe, curriculum-aligned path - it builds on what is already in the Australian Curriculum rather than bolting on something new, and it aligns the school's effort with where the labour market is heading.

Why this matters now

Schools are under pressure to respond to AI, and the safest, most durable response is to extend an existing capability rather than invent a new one. The demand signal is unambiguous. The World Economic Forum's Future of Jobs Report 2025 names AI and big data the fastest-growing skill, sitting just behind analytical thinking - the most important core skill overall - with technological literacy also rising and 39% of workers' core skills expected to change by 2030. PwC's adjacent labour-market work has documented a 56% wage premium for AI-skill roles; the capability students build at school increasingly determines their footing after it.

The Australian evidence is what makes this concrete for a school council weighing where to spend scarce curriculum time. Jobs and Skills Australia's 2025 study Our Gen AI Transition - the first whole-of-labour-market analysis of the technology in Australia - finds generative AI augments more roles than it replaces and lifts demand for digital literacy alongside human skills such as problem-solving, communication and adaptability. That is not an argument for narrow technical training; it is an argument for exactly the judgement digital literacy was redesigned to build. The macro stakes give the point weight: the Productivity Commission's August 2025 interim report estimates AI could add roughly $116 billion to GDP over a decade and lift labour productivity by about 4.3%, while McKinsey's State of AI 2025 finds 88% of organisations already use AI but only around 7% have fully scaled it. The lesson from industry is the one schools should internalise - access is easy, capability is hard, and value accrues to those who can actually use the technology well. A school that teaches AI literacy as disciplined judgement is preparing students for the half of the economy that captures the upside, not the half still struggling to.

At the same time, the policy and curriculum scaffolding to do this responsibly already exists in Australia. The Australian Framework for Generative AI in Schools - six principles spanning Teaching & Learning, Human & Social Wellbeing, Transparency, Fairness, Accountability, and Privacy, Security & Safety, supported by 25 guiding statements - was approved by Education Ministers in October 2023, and its 2024 review was endorsed again by Ministers in June 2025. Schools do not need to start from a blank page. They need to recognise that AI literacy is a layer on a foundation they have already been building, and our explainer on what AI literacy means for Australian students sets out that layer in detail.

What digital literacy already established

Digital literacy is more demanding than the label suggests, and that is precisely why it makes such a strong foundation for AI literacy. In Australian Curriculum Version 9.0, ACARA renamed the ICT Capability general capability to Digital Literacy - and the rename was substantive, not cosmetic.

The shift broadened the capability in three ways that matter directly to AI:

• From how to why and when. ICT Capability leaned toward operating tools. Digital literacy added the judgement of why to use a technology and when it is and is not appropriate - the same judgement that decides whether reaching for AI helps or harms a piece of learning.
• A sharper focus on privacy, security and online safety. Digital literacy put protecting yourself and others at the centre, not the periphery. AI raises every one of those stakes, from what students feed into a tool to what they trust coming out.
• Critical evaluation of digital content. Students were already expected to question what they found online. AI simply makes that skill non-negotiable, because the content is now generated on demand and arrives without a source.

In other words, digital literacy already taught students to be purposeful, safe and critical with technology. That is most of the disposition AI literacy needs. The new layer is not a new mindset - it is the specific knowledge and habits that a generative technology demands. And the broader international picture confirms Australia is building on a real footing rather than improvising: Stanford HAI's AI Index 2025 reports that two-thirds of countries now offer or plan K - 12 computer-science education, double the share in 2019, even as it warns that teacher readiness lags intent. Australia's curriculum shift put the foundation in place ahead of many systems; the task now is to extend it deliberately.

Where AI literacy extends the foundation

AI literacy is digital literacy applied to a technology that behaves unlike any tool that came before it. A search engine retrieves; a generative model produces. That single difference is what the new layer addresses.

Three extensions matter most, and each maps to an established competency framework rather than to opinion:

• Understanding how AI fails. Digital literacy taught students to evaluate sources. AI literacy adds the prior question: this output has no source - it was predicted, it may be invented, and it is delivered with total confidence regardless of accuracy. UNESCO's AI Competency Framework for Students (2024) names exactly this under its "AI techniques" dimension: a student needs enough understanding of the machine to judge it.
• Directing, not just retrieving. With a search engine, the skill was finding. With AI, the skill is directing - giving context and constraints, framing a problem so the output is genuinely useful. This is a distinct competency that digital literacy never had to teach, and it is the one the labour market is pricing most directly.
• A new kind of integrity question, governed by values. Digital literacy worried about plagiarism and safe sharing. AI literacy adds a subtler line: when does using AI cross from honest help into quietly outsourcing the thinking the task was meant to develop? UNESCO frames this under "ethics" and a "human-centred mindset"; the OECD's AI Principles (2019, updated 2024) provide the values layer beneath it, defining what trustworthy, human-directed use looks like. That is the question explored in depth in using AI versus learning with AI.

None of this displaces digital literacy. It sits on top of it, which is exactly why it is school-safe to introduce - there is no new mindset to install, only a new layer to add to a familiar one.

How AI literacy builds on digital literacy in practice

The continuity is clearest in concrete classroom work, where the digital-literacy skill and the AI-literacy extension appear side by side. Three examples show the layering.

• Evaluating a source becomes evaluating a generated claim. A student researches a topic and, where digital literacy taught them to weigh a website's credibility, now must also verify an AI summary that cites nothing. What the student does today: gathers background for an assignment. How AI assists: it produces a fast, readable overview. What the student must verify: every claim against a real, attributable source. The learning outcome: critical evaluation extended to sourceless content. The control: the AI overview is a starting point, never a citation.
• Online safety becomes data judgement. A student who learned not to overshare on social platforms now applies the same instinct to what they paste into an AI tool. What the student does today: decides what to put into a prompt. How AI assists: it can work with whatever it is given. What the student must verify: that nothing private or sensitive - theirs or anyone else's - goes in, in line with the framework's Privacy, Security & Safety principle. The learning outcome: privacy judgement extended to a new surface. The control: the student decides what the tool never sees.
• Purposeful tool use becomes deliberate direction. A student who learned to choose the right digital tool for a task now learns to direct that tool with context and constraints. What the student does today: sets out to produce something specific. How AI assists: it responds far better to a clear brief than a vague one. What the student must verify: that the result is correct, not merely confident. The learning outcome: purposeful use deepened into deliberate command - the capability McKinsey's data suggests separates organisations that scale AI from those that merely adopt it. The control: the student sets the brief, not the machine.

Each example is one capability with a new layer - which is exactly why schools can teach it without dismantling anything.

How schools can layer AI literacy on digital literacy

Schools do not need a curriculum overhaul to introduce AI literacy. They need to treat it as the extension it is.

1. Anchor it to digital literacy that already exists. Position AI literacy explicitly as the next layer, so teachers and students see continuity rather than a new and unfamiliar demand. ACARA's Version 9.0 framing is the natural anchor point.
2. Lead with the framework's safety principles. Privacy, Security & Safety is already one of the six principles in the Australian Framework - make it the entry point, because it is where digital literacy and AI literacy overlap most cleanly, and it carries clear ministerial endorsement.
3. Teach the new failure modes explicitly. Students need to be shown why AI invents facts and serves them confidently, because that is the part digital literacy never had to cover. UNESCO's progression treats this understanding as foundational, not advanced.
4. Build the metacognitive habit, not just the technique. The evidence here is unusually strong: the Education Endowment Foundation, working with Evidence for Learning in Australia, finds that teaching metacognition and self-regulated learning is among the highest-impact, lowest-cost strategies available, worth roughly seven months of additional progress when taught alongside subject content. AI literacy is, at its core, a metacognitive discipline - thinking about your own thinking before, during and after you reach for the tool - which is why it belongs woven through learning areas rather than bolted on as a one-off.
5. Build teacher capability first. The capability gap is often at the front of the room; teachers cannot extend a literacy they have not been supported to build themselves. Stanford HAI's finding that fewer than half of computer-science teachers feel equipped to teach AI, despite 81% believing it belongs in foundational education, is the clearest signal that this step cannot be skipped.
6. Use structured programs for depth. Curriculum capabilities set the floor; deliberate, sequenced AI education adds the depth, which is what our AI literacy programs are designed to provide.

Common mistakes

• Treating AI literacy as a brand-new subject. Framed as separate, it competes for space and feels optional; framed as a layer on digital literacy, it is curriculum-aligned and obviously necessary.
• Skipping straight to tools. A login is not a literacy. Without the judgement layer, students just use AI faster and less safely - the gap between McKinsey's 88% who adopt and the ~7% who scale, recreated in miniature in the classroom.
• Assuming digital literacy already covers it. The disposition transfers; the specific knowledge of how generative AI fails does not.
• Upskilling students while leaving teachers behind. The capability gap simply moves to the adults guiding the work - precisely the readiness gap Stanford HAI documents.
• Ignoring the safety dimension. AI raises the privacy and security stakes digital literacy already flagged - treating it as purely an academic skill misses half the point and sidesteps the framework's clearest principle.

How to know the layering is working

It is working when students treat AI the way well-taught digital literacy taught them to treat all technology: purposefully, safely and critically - only with the sharper judgement a generative tool demands. They check a sourceless AI claim the way they would weigh a dubious website. They guard what they feed a tool the way they guard what they post. They direct, rather than passively retrieve. The failure mode is the opposite: students who are comfortable with the interface but have no judgement about the output, mistaking fluency for literacy. For the full picture of how this fits into a young person's education, see our guide to AI education for teenagers in Australia.

The recommendation for schools is clear. Do not stand up AI literacy as a separate, competing subject, and do not assume digital literacy already covers it. Treat AI literacy as the natural next layer on the digital literacy your curriculum has been building since Version 9.0 - anchor it to the safety principles you already teach, name the new failure modes that generative AI introduces, lean on the metacognitive habits the evidence rewards, and build teacher capability before student capability. The commercial case reinforces the pedagogical one: in an economy where Jobs and Skills Australia expects AI to augment most roles and lift demand for human skills, and where value accrues to those who can genuinely use the technology rather than merely access it, the school that layers AI literacy onto digital literacy is doing the most durable thing it can with the time it has. Done this way, AI literacy is not a disruption to the curriculum. It is its logical continuation.