Two Postcodes, Two AI Educations: The Digital Divide in the Classroom

by | Sep 1, 2026 | Education in the Age of AI

Skip to main content
< All Topics
Updated
ByThe Good

Every other article in this series treats "AI in schools" as one national story. It isn't. It's several different stories happening at once, depending on which school a student happens to attend — and the gap between them is arguably the least discussed, most consequential part of the whole debate. AI doesn't create educational inequality in Australia; it lands on top of inequality that already exists, and there's a real risk it widens it faster than any policy currently in place can close it.

The gap is already visible before AI enters the picture

Even before AI is added to the equation, the numbers are stark. Within Sydney alone, only 31% of students in Western Sydney have access to a digital device for learning at home, compared with close to 60% across the rest of the city — a divide inside a single metropolitan area, not just between cities and the bush. Nationally, more than one in five students in schools where disadvantage is common lack basic digital resources, compared with roughly one in fifty students in wealthier schools. That's the foundation AI-powered learning is being layered onto: a device and connectivity gap that predates the current AI wave by years and hasn't closed.

Regional and remote schools face a second, compounding gap

Layer AI on top of that and rural and remote communities face the sharpest version of the problem — unreliable internet and limited technology access make AI-powered learning tools genuinely hard to deploy, on top of the same regional teacher shortages the workforce article in this series has already covered, and which fall hardest on First Nations communities. Gunbalanya School, a remote school in West Arnhem Land in the Northern Territory, offers a useful case study in both directions: despite unreliable internet and uneven staff IT skills, it has made real progress on children's literacy with digital tools — proof this isn't an unsolvable problem — but that progress has come through sustained, targeted external support rather than anything close to what's available by default in a well-resourced metropolitan school.

Private schools are moving faster than public ones

The adoption gap isn't only geographic — it's also sectoral. Surveys of independent schools report more than 70% of primary and over 80% of secondary teachers in some of those schools already experimenting with AI, well ahead of average uptake across the public system. Well-resourced schools have the budget for staff training, procurement of vetted tools and dedicated IT support; public schools in lower-SES areas are, in many cases, still working on foundational digital inclusion before AI literacy becomes a realistic next step. That's not a story about willingness — it's a story about which schools have the spare capacity to experiment safely in the first place, and it echoes almost exactly in an unexpected place: NSW's new cap on take-home HSC assessments (covered in the assessment article in this series) shifts more weight onto supervised, in-class work — which assumes every school has the timetabling flexibility, invigilation staff and classroom capacity to absorb that shift equally. A well-resourced independent school and a stretched public school in a growth corridor are not starting from the same place to comply with the same rule.

Government is aware of the problem — implementation is another matter

To be fair to policymakers, the equity risk isn't unrecognised. The federal government's National AI Plan explicitly commits to spreading AI's benefits equitably, upgrading infrastructure, funding digital inclusion programs and respecting First Nations data sovereignty. Some targeted pilots are already running — Western Australia has a $4.7 million jointly federal- and state-funded AI pilot exploring how AI can cut teacher workload through lesson planning and admin support, one of a growing number of state-level programs. But education researchers watching the rollout have made the same point repeatedly: schools in regional, remote or low-SES areas are already affected by the digital divide today, and infrastructure and digital-inclusion commitments that arrive on a multi-year timeline don't help a school that needs to make decisions about AI use this term. A national plan and a state pilot program are not the same thing as reliable internet and enough devices in every classroom right now.

South Australia's royal commission puts equity formally on the table

The most significant recent development here isn't a funding announcement — it's an inquiry. On 10 August 2026, South Australian Premier Peter Malinauskas announced Australia's first royal commission into artificial intelligence, commencing 1 October 2026 with a report due to government by 1 July 2027. Its terms of reference explicitly cover AI's effect on the state's secondary and tertiary education systems, alongside the economy, public services and the creative sector — making it the first formal government inquiry in the country to treat education equity as core AI policy business rather than an afterthought bolted onto a curriculum announcement. Malinauskas has framed the commission as proof that "we can choose to set the agenda" rather than let AI adoption happen to the state unmanaged — a notably different posture to the reactive, integrity-driven urgency behind NSW's assessment reform, covered in this series' second article. Whether an inquiry that reports in mid-2027 can move fast enough to matter for students already navigating today's digital divide is, fairly, an open question — and one the commission itself will have to answer.

What a fix would actually require

The most concrete equity proposal circulating in 2026 outside the South Australian inquiry is a "Digital Equity Learning Guarantee" — free or low-cost access to devices and connectivity for disadvantaged students, dedicated funding to lift digital and AI literacy skills, and professional learning for teachers so uptake isn't limited by staff confidence as well as student access. Researchers are also clear the fix isn't only about hardware: genuine equity in AI use means tools built around local curricula, respectful of First Nations knowledge systems, and available in accessible formats — not just a laptop rollout with a generic chatbot bolted on.

The question worth sitting with

If AI fluency becomes as important to future employability as literacy and numeracy already are — which is the explicit bet behind the national AI literacy curriculum — then a digital divide that starts with who has a laptop at home risks compounding into something with permanent economic consequences. Is closing that gap a genuine national priority government funding will match on the timeline students actually need, or a well-intentioned commitment — and now a two-year royal commission — that stays several years behind the technology and the assessment reforms it's meant to keep pace with?

Sources

Table of Contents