Why workplace literacy is the foundation of every AI strategy in New Zealand
- 2 days ago
- 6 min read
Every business publication, LinkedIn feed, and Monday morning leadership meeting is talking about the same thing right now: AI. Get your team AI-ready. Roll out Copilot. Train your managers in prompt engineering. Don't get left behind.
There's nothing wrong with the message. The opportunity is real. Research by Accenture for Microsoft estimates generative AI could contribute between $76 billion and $108 billion annually to the New Zealand economy by 2038.
CEOs across the country are already seeing it. PwC's 2025 CEO Survey found 70 percent of New Zealand CEOs say AI has increased efficiencies in their employees' time at work, significantly higher than Australia (42 percent), the Asia-Pacific region (58 percent), and globally (56 percent).
But here's the part nobody's saying out loud.
You cannot build AI fluency on top of shaky foundations. And right now, in production lines, warehouses, kitchens, and admin offices across New Zealand, the foundations are shakier than most leadership teams realise.
The Hidden Prerequisite
AI literacy isn't a standalone skill. It sits on top of three things your frontline workers need first.
Reading. Following written prompts, interpreting AI-generated outputs, spotting when the answer is wrong.
Numeracy. Sense-checking calculations, understanding data, knowing whether a figure is plausible.
Digital confidence. Logging into a system, navigating a web interface, knowing what a browser tab is.
If a team member struggles to read a production schedule, they're not going to suddenly thrive with an AI assistant that generates one. If they panic at a spreadsheet, asking them to validate Copilot's output is a fantasy. If they avoid email because typing is hard, no amount of Microsoft licensing is going to change their working day.
The OECD's 2023 Survey of Adult Skills (PIAAC) made the scale of this plain. In New Zealand, 26 percent of adults scored at Level 1 or below in literacy, and 28 percent scored at Level 1 or below in numeracy. Over half the adult population (51 percent) scored below Level 3 in literacy, and 60 percent scored below Level 3 in numeracy. Compared to 2014, literacy fell by 21 points and numeracy by 15 points, both far exceeding the average OECD decline of six points. Ako
That's the workforce most New Zealand employers are about to drop AI tools onto.
The New Zealand Paradox
Look at how this is playing out in 2026 and the contradiction is striking.
At the top: 87 percent of New Zealand companies say job roles have changed or disappeared because of AI in the past 12 months, and 67 percent are investing in AI training and reskilling programmes.
At the frontline: A global study by KPMG and the University of Melbourne found New Zealanders are among the least confident and least trained when it comes to using AI. Only 41 percent of Kiwi workers say they use AI at work, compared to 91 percent in India. The same study found 76 percent of New Zealanders had received no formal or informal AI training, and more than 60 percent said they didn't feel confident using it.
That gap between strategy and shop floor is where AI investments quietly fail. A leadership team buys the licences. A consultant runs a half day workshop. Everyone leaves enthusiastic. A month later, adoption is minimal, because the workflows didn't change, the supervisors didn't model it, and a meaningful slice of the team can't comfortably read the interface in front of them.
McKinsey's research on AI adoption is blunt: training alone almost never drives sustained behaviour change. What drives it is people having the underlying capability to engage, leadership going first, and the skill building being embedded in daily work.
In other words: start with the foundations.
What AI-ready actually looks like on the factory floor
We recently completed a workplace literacy and numeracy programme with an Auckland manufacturer, a mid-sized natural personal care company with a frontline production team. Like most New Zealand manufacturers, they had a workforce mix: long-tenured, hands-on operators alongside newer staff. English wasn't always the first language. NCEA pathways had been patchy.
When we started, packaging and labelling errors were running at 15 to 20 incidents per week. Staff weren't filling out health and safety forms. Digital tools sat unused. Teams, email, even basic Excel were avoided rather than used.
We didn't run an AI training programme. We ran an EWLN programme, Employer-led Workplace Literacy and Numeracy, funded through the Tertiary Education Commission. We tied every learning task to a real workplace problem: reading a batch sheet, calculating fill volumes, completing an H&S form, sending a Teams message.
Twelve months later:
Packaging and labelling errors dropped from 15 to 20 a week to roughly one a fortnight, a 95 percent reduction.
Numeracy assessment scores (LNAAT) climbed 96 percent on average across the cohort.
Reading scores climbed 31 percent.
Health and safety forms, previously the domain of one or two senior staff, were being completed across the whole team.
And here's the part most relevant to this article: staff started using internal digital software, email, and Excel daily. Without us teaching any of those tools as the primary objective.
The foundational skills unlocked the digital ones. Now those staff are ready for the next conversation, the one about Copilot, AI process automation, and using ChatGPT to draft customer responses. A year ago, that conversation would have landed in a vacuum. Now it lands on solid ground.
That's what an AI-ready frontline workforce actually looks like in New Zealand. Not a glossy training video. A team that can read, calculate, and navigate confidently, and is therefore capable of learning what comes next.
The two-track approach
The smartest employers we work with are running two tracks in parallel, and getting government funding for both.
Track one targets the team members who've been quietly avoiding written tasks, digital systems, and anything involving numbers. The goal is foundational confidence: reading, numeracy, basic digital fluency. This is the group AI rollouts always forget about, and it's usually 20 to 30 percent of any frontline workforce.
Track two targets the team members who already have foundations and are ready to accelerate. Excel skills, problem solving, AI tools, process improvement, communication. This is where you build the early adopters and the eventual supervisors, the people who'll champion AI tools because they actually understand what the tools are doing.
Both tracks can be funded under the same EWLN application. Both contribute to the same productivity outcome. Neither one works on its own.
The funded on-ramp most employers haven't used
Here's what catches most leadership teams off guard: this is funded.
The TEC's Employer-led Workplace Literacy and Numeracy fund covers the cost of programme design, delivery, and assessment for eligible New Zealand employers. Application deadlines for 2026 fall in February, April, June, August, and October. Five chances a year for a New Zealand employer to put their hand up.
Most don't. The fund is underused, partly because employers assume it's only for "remedial" workforces, and partly because the application process feels opaque. Both assumptions are wrong. The fund is designed for the exact scenario most New Zealand businesses are facing right now: an existing workforce whose skill ceiling is becoming a productivity ceiling, just as AI is raising the bar of what every job demands.
If you're an Operations Manager, GM, or business owner asking "how do we get our team AI-ready?", start one step earlier. Ask: do my people have the foundational reading, numeracy, and digital skills to engage with what's coming?
If you don't know the answer, that's the answer.
The real competitive advantage
The businesses that win the next decade in New Zealand won't be the ones who deployed AI fastest. They'll be the ones who built workforces capable of using it. And that capability starts well before the first AI tool gets switched on.
AI doesn't care that your team can't read the schedule. It will simply produce outputs they can't validate, prompts they can't write, and gains they can't capture.
Foundations first. Then everything else.
Book a FREE Assessment
At Zeducation, we work with New Zealand employers to design and deliver TEC-funded EWLN programmes that build the foundations your team needs to thrive in an AI-enabled workplace. If you want to know whether your workforce is ready, or what a programme could look like, let's talk.
Reach out for a free assessment of your team's literacy and numeracy needs.
Comments