Quick-reference summary - Current state: AI is in classrooms and admin systems across Australia in 2026 — from NSW primary schools to remote high schools. Policy patchwork across states means practice varies. - Key dates: NSW issued formal guidance in 2024; Victoria introduced a statewide policy in 2025; by mid‑2026 every state and territory had either policy or mandated guidance for government schools. - Common uses: lesson planning, differentiation, marking support, student wellbeing checks, enrolment and attendance analytics, and facilities maintenance scheduling. - Typical tools: ChatGPT (OpenAI) — ChatGPT Plus ~£16/month; Canva for Education (Canva added AI features in 2023–24); Microsoft Copilot features in M365 Education; Google Duet for Workspace for Education. - Main concerns: assessment integrity, data residency, internet bandwidth in regional areas, equity of access, and teacher training.
Overview — the current state in 2026
AI isn't a classroom fad any more. In 2026 teachers use AI for lesson planning, differentiation, marking support and wellbeing checks. Students use generative tools for drafting essays, getting personalised practice and building multimedia projects. Schools also use AI behind the scenes — enrolment systems, attendance analytics and maintenance scheduling are increasingly automated.
It's messy — different states and territories are handling AI in very different ways. Some systems have moved fast to adopt vendor services; others are cautious. Parents and teachers want clarity. Students want skills. Administrators want to reduce workload. Right now the big story is the clash between AI's clear benefits and the real risks schools face.
On the ground, that looks different depending on where a school sits. Metropolitan government primary schools that run Google Workspace or Microsoft 365 often have district‑level contracts that include AI features. Independent schools tend to trial commercial edtech bundles. Remote and very small schools are slower to adopt — not just because of cost, but because of connectivity and local IT capacity. Still, even small schools report using AI for administrative automation — for example, automated rostering and asset management linked to cloud systems.
Key developments since 2023
Several things shaped uptake into 2026. First, big vendors made education‑specific products. Microsoft pushed Copilot features into Microsoft 365 Education from 2023–24; Google expanded Duet AI for Google Workspace for Education across 2024; and OpenAI introduced classroom‑friendly modes for ChatGPT in late 2023 and iterated them through 2025.
Those changes let teachers use AI inside the apps they already open every day.
Second, Australian jurisdictions started publishing formal guidance. NSW issued formal guidance in 2024 and Victoria followed with a statewide policy in 2025. By mid‑2026 every state or territory had either policy or mandated guidance for government schools — so advice now exists in every education department for public systems. Those documents vary: some ban covert use during summative assessment, others focus on consent and data handling, and a few include curriculum updates for digital and AI literacy.
Third, local industry stepped up. Australian edtech firms and platforms — including Canva, which rolled out several AI features in 2023–24, and smaller local curriculum platforms — offered integrations and claimed local data handling. Research organisations such as CSIRO's Data61 continued to advise on trusted AI and privacy. Some vendors started offering hosting in Sydney-based cloud regions to ease schools' concerns about where student data sits. That simple change pushed more schools into pilot projects because state procurement often requires clarity on where student data is stored.
Top picks — tools teachers actually use (with UK price comparisons)
Below are tools most commonly trialled or deployed in Australian schools, with typical UK retail prices for comparison — many schools buy through education agreements at lower per‑user rates or via district licences.
- ChatGPT (OpenAI) — free tier available; ChatGPT Plus in the UK: ~£16/month. Used for drafting prompts, generating formative questions, and quick content checks.
- Canva for Education — Canva is Melbourne‑based and rolled out AI writing and image tools in 2023–24; Canva Pro is around £11.99/month in the UK for individuals, while Canva for Education is free for many schools under a site licence.
- Microsoft 365 Education with Copilot features — schools on enterprise agreements can add Copilot capabilities; Microsoft announced Copilot pricing for business tiers in 2023 and education deals vary by contract (often negotiated at state or system level).
- Google Workspace for Education with Duet AI — Google expanded Duet in 2024; education tiers are commonly procured at system level and prices vary by contract.
- Local LMS and assessment tools — several Australian curriculum platforms now include AI‑assisted marking or feedback; pricing is mainly by school licence and negotiated regionally.
In practice, most schools pay far less than retail prices because they buy through system-wide agreements. State procurement and multi‑school consortia often reduce per‑user costs by 50% or more. That makes otherwise expensive AI features affordable for larger public systems.
Industry impacts — assessment, workforce and equity
The most heated debates are about assessment and how to keep summative tests fair. Formative use of AI is widely accepted — teachers lean on tools for personalised practice and feedback. But for summative assessment, many systems now forbid unsupervised use. Some jurisdictions require teacher oversight and explicit student declaration when AI is used in assessment tasks. According to the article, initial guidance from 2024 hardened into stricter rules in some states by 2025–26.
Basically, teacher workload is changing in two ways. AI cuts repetitive tasks — draft marking rubrics, generate reading lists, format worksheets — but it adds new tasks: checking AI outputs, reworking prompts, and teaching digital literacy. Schools that invest in coordinated professional learning report smoother rollouts. Uncoordinated ad hoc use, by contrast, can increase workload and cause friction.
Equity is another headline. Schools in wealthier suburbs or well‑resourced systems get earlier access to paid tools and better internet connections. Regional and remote schools often rely on lighter‑weight or offline strategies. The national broadband picture helps here: the presence of NBN and mobile coverage varies, and that affects whether a school can realistically run heavy cloud models in real time.
Expert views — what educators and researchers are saying
Educators and academics are cautious but pragmatic. Curriculum experts say students need explicit instruction in AI literacy — how models work, what bias looks like and how to verify facts. Teacher leaders emphasise assessment design: craft tasks that test reasoning and process, not just final text that an AI can produce.
Data and privacy specialists sound the alarm on contracts and data flows. They're pushing for clear vendor obligations on student data, with many districts insisting on Australia‑based hosting and contract clauses that forbid reuse of student submissions to train external models. Procurement teams report that those clauses are now common bargaining items in 2025–26 tenders.
Union and parent groups want transparency. Teacher unions raise workload and integrity concerns. Parent groups press for clear consent and opt‑out pathways for younger students. Governments are responding with guidance and, where needed, regulation — which explains the wave of policy documents across states in 2024–26.
What's next — short‑term moves and longer trends
In the short term — the next 12–24 months — expect more focus on training and governance. Education departments will push system‑level professional learning and standard vendor contract clauses. Some states are likely to centralise procurement of AI services to control costs and data flows.
Longer term, AI will change what schools teach. Digital literacy and AI ethics are being folded into curriculum updates in several jurisdictions. Schools will need to teach how to work with AI — prompt design, critical evaluation of machine outputs and the ethics around data sharing. That won't happen overnight, but by the late 2020s these skills will be a normal part of secondary courses in most systems.
Finally, expect more hybrid models: local lightweight inference for privacy‑sensitive tasks, and cloud models for heavy content generation. Vendors already offer Australia‑region hosting and on‑prem or private‑cloud options. That trend will make adoption less risky for cautious systems.
Related Articles
School leaders who move early and sensibly — with clear policy, chosen platforms, teacher training and strong vendor contracts — will get the most from AI while keeping students safe. By mid‑2026 the balance looks like this: steady adoption in metro systems, piecemeal use in independents, and cautious pilots in regional and remote schools. The debate will keep shifting from "can we use AI" to "how do we teach students to use it well," and that's where most effort will land over the next few years.
This article was created with AI assistance.
