By Susana Tomaz – Director of Futures and AI Strategy, Westlake Girls High School, International Research Centre for Artificial Intelligence (UNESCO) Fellow, Global Mentor on AI in Education for Asia Europe Foundation, Developer of Open Education Resource
I’ve spent a lot of time in classrooms,learning design workshops, staffrooms, and leadership meetings over the last 20 years in education. Not in the abstract sense of “education is changing,” but in the very real sense of watching teachers navigate new expectations, new tools, and new pressures — alongside the same immovable constraints of time, workload, and responsibility for young people.
If I had to summarise what I’m seeing across New Zealand schools right now, it would be this:
The system is being asked to change faster than its capability to change is being built.
Schools are simultaneously navigating curriculum refresh, workforce shortages, and wellbeing pressures, alongside growing assessment uncertainty. Just as questions about how we evidence learning are intensifying, AI tools are rapidly reshaping how work is produced. With agentic browsers now capable of completing assignments and navigating online assessment platforms on behalf of students, the foundations of independent assessment are being fundamentally challenged.
Education systems around the world are trying to secure academic integrity while the technological landscape keeps shifting beneath them.
When we focus only on policing, we stay reactive.
When we invest in policy, pedagogy, and practice, we become intentional.
The biggest risk isn’t cheating. It’s cognitive offloading.
A large US follow-up study of more than 4,000 high school students found that overall cheating rates remained consistent with historical baselines after AI became widely available. What changed wasn’t dishonesty — it was how students used AI.
Most reported using it for idea generation, explanations, and drafting support.
The OECD’s Digital Education Outlook 2026 describes the “crutch effect”: students given unrestricted AI access performed 48% better during practice, but 17% worse on independent exams. When AI removes productive struggle, performance can rise while learning declines.
But from the frontline, the real challenge isn’t the technology. It’s capability, confidence, and coherence.
For nearly 200 years, schooling has leaned heavily on the lower levels of Bloom’s Taxonomy: remembering, understanding, applying. But AI now performs many of those tasks instantly.
As Ron Beghetto warns in the OECD report:
“If education is just about delivering inert content for students to reproduce, machines will do that better.”
If education stays anchored here, students will always trail behind machines.
We need to lean deliberately into the upper levels: analysing, evaluating, creating. The OECD warns that overreliance on AI tools that provide direct answers can reduce active engagement and weaken deep learning if pedagogy is not intentionally redesigned.
This is the real disruption.
If education is about reproducing information, AI wins.
If education is about meaning-making, creativity, and ethical reasoning, humans still lead.
The question is whether we are brave enough to redesign learning around what only humans can do well.
The OECD is clear: generative AI can enhance learning and productivity, but only when human judgement and oversight remain central. This is not a tool problem. It is a governance problem.
At Westlake Girls’ High School, we’re not trying to stop students from using AI. Our role is to help them engage with it critically, ethically, responsibly, and in ways that strengthen — not replace — their thinking.
The gap between adoption and capability is where the real risk sits.
Internationally, countries moving with foresight show consistent patterns:
- Clear national, system-level strategy
- Explicit links to curriculum and learning design
- Strong investment in teacher capability
- A focus on ethics, safety, and wellbeing
- A commitment to equity and access
Yet New Zealand’s first national AI Strategy focuses on productivity and competitiveness, while education remains largely absent from its centre. The EU’s AI Act, by contrast, explicitly embeds AI literacy as a workforce priority.
If education is missing from our AI strategy, we are outsourcing our future workforce capability.
Education is how we grow our local talent pool. It is how we build the cognitive capability, digital fluency, and ethical reasoning required for an AI-enabled economy. If we are serious about lifting productivity, meeting GDP targets, and reducing reliance on imported expertise, AI readiness cannot stop at industry.
Children entering secondary school this year will enter the workforce around 2031 — into an AGI-shaped world. AI readiness must begin in schools.
In my March 2024 EdTechNZ blog, I called for a national AI framework for New Zealand schools. Sixteen months on, we still lack a coherent, education-specific framework to guide safe and equitable implementation.
Without coherent national direction, individual schools are left to manage risk, build capability, and prepare the next generation alone.
Without coordinated national action, AI risks compounding existing inequities.
If we want meaningful, responsible innovation in education, we need to stop treating schools as the place where change lands and start treating them as the place where change is deliberately designed, supported, and built.
The biggest risk to students isn’t AI. It’s that our systems haven’t yet caught up to support teachers and leaders to guide them well. And this is exactly why initiatives like Day of AI Aotearoa matter.
Day of AI Aotearoa| New Zealand is an AI literacy programme, connecting local classrooms to the global MIT RAISE Day of AI movement. It ensures students develop critical, ethical, and creative understanding of artificial intelligence by localising globally recognised, research-informed resources to align with Aotearoa’s curriculum and context.
The initiative is delivered in collaboration with MIT RAISE and the global Day of AI team, in partnership with Day of AI Australia, and supported by Technology Education New Zealand (TENZ), the Education Partnership & Innovation Trust (EPIT), and NZCER, which evaluated the 2025 pilot.
“The kids no longer saw AI as this all-knowing thing …. and that changed everything.”
— Teacher, pilot school
“I went from feeling like a two out of ten to a six or seven. I now feel confident having these conversations with my students.”
— Teacher, pilot school
Following a successful Term 4 pilot across ten schools, independently evaluated by NZCER, the programme is rolling out nationally on 9 March. It’s designed to work with the realities of schools.It can run as: a whole-school Day of AI, a sequence of 4 lessons, 8 flexible 30-minute blocks or integrated across learning areas. That means schools can choose what fits their timetable, their students, and their learning priorities, when it works for them.
By Term 1 2026, every school, teacher, and learner in Aotearoa will have access to free, high-quality AI literacy.
We are shaping the future today. The more AI literacy we build in our young people, the more agency they will have to make good choices as a society. And they need to be part of that conversation, because the future is theirs.
The future is not a destination we go to. It is something we create.
