New Zealand now has an official map of where AI is likely to touch its workforce — and it is both calmer and more pointed than the international noise. A May 2026 briefing to the Labour Market Ministers Group, released to Stuff under the Official Information Act, found only 2% of current NZ occupations are at high risk of full automation with AI at today’s capability levels, chiefly software programmers, ICT professionals, accountants, and call-centre operators.
🔍 THE BOTTOM LINE — The more consequential number is the 36% of occupations with high potential for AI augmentation, concentrated among managers and knowledge workers. For most Kiwi workers, the realistic scenario is not replacement this year — it is a changing job description, and the workers best positioned are those whose employers take augmentation seriously.
What the Briefing Says
The MBIE modelling reported by Stuff makes three points worth holding together:
- Full automation stays rare. At 2% of occupations, “scenarios of AI leading to rapid displacement and mass unemployment appear unlikely at this stage,” the briefing stated.
- Augmentation is the real story. 36% of occupations show high potential for AI to assist the work — and that group is weighted toward managers and higher-skilled knowledge workers, not the factory floor or the paddock.
- Churn is normal. New and disappearing jobs already account for 13 to 15% of all roles in a typical year. Officials’ position is that AI gets absorbed into churn that the labour market handles every year anyway.
That official calm sits alongside a second data point from the same briefing, though. AI Forum data cited by MBIE found 45% of surveyed NZ businesses planned to lower headcount as they implement AI. Modelling says gradual absorption; boardroom surveys say restructuring. Both can be true — displacement through attrition and retirement rather than abrupt layoffs — but the gap between them is where individual workers’ risk actually lives.
Who Is Exposed: The RBNZ Picture
The Reserve Bank’s February 2026 analytical note, “Assessing AI and Robotics Exposure in the New Zealand Labour Market Using Large Language Models”, scored every NZ occupation using ten different LLMs — ChatGPT, Claude, DeepSeek, Gemini, GLM, Grok, Kimi, Minimax, Perplexity, and Qwen. Its findings:
- Around 30% of NZ workers face high combined exposure to AI and robotics. Fewer than 5% face low exposure to both, meaning almost everyone will engage with at least one of these shifts.
- AI exposure concentrates in professional, managerial, and administrative occupations — and in the $50,000 to $150,000 income band. That is the middle layer of most NZ organisations.
- Robotics tells the inverse story, concentrating in machinery operator and labourer roles, and hitting workers with no qualifications harder than those with tertiary credentials.
Two conclusions follow. First, the exposure map inverts the old automation story: this wave points at the office, not the workshop. Second, since AI exposure lands hardest on mid-band white-collar roles, the augmentation half of the picture matters most — a 36%-of-occupations group that includes the people who run, staff, and manage NZ businesses.
The Positive Case (It’s Not Just Hedging)
The same global evidence base that warns about augmentation also rewards it. PwC’s 2026 AI Jobs Barometer found jobs requiring specific AI skills growing far faster than the total jobs market, with an average wage premium for AI skills of around 62%, and its New Zealand edition reported AI-related job ads nationally rising from roughly 3,900 in 2024 to 9,600 in 2025. Manufacturing emerged as the surprise NZ leader, with an AI wage premium of about 62% — ahead of every other sector.
MBIE’s own framing supports a managed transition rather than a cliff. If AI absorbs into normal churn, the differentiator for workers is not avoiding exposure — fewer than 5% manage that — but being on the augmented side of it: roles where the AI handles the codifiable 60% and the human judgement, client relationships, and site decisions become the scarce, better-paid part. PwC New Zealand’s June report put it as NZ’s opportunity being to become “one of the world’s smartest adopters” — applying AI to productivity rather than trying to outspend larger economies on infrastructure.
FAQ
What percentage of NZ jobs could AI fully automate? MBIE’s May 2026 briefing put it at 2% of current occupations at today’s capability levels — mainly software programmers, ICT professionals, accountants, and call-centre operators.
Which NZ workers are most exposed to AI? RBNZ research using ten LLMs found AI exposure concentrates in professional, managerial, and administrative roles in the $50,000–$150,000 band. Robotics exposure is the opposite — machinery operators and labourers — and around 30% of workers face high combined exposure.
Is AI hiring growing in New Zealand? Yes. PwC’s AI Jobs Barometer found AI-related NZ job ads rose from roughly 3,900 in 2024 to 9,600 in 2025, with AI skills appearing in 2.7% of all job postings, and NZ manufacturing showing an AI wage premium of around 62%.
NZ Angle
This story is the NZ angle — it’s the country’s own officials modelling its own labour market, from MBIE’s ministerial briefing to the Reserve Bank’s exposure scoring. For Tauranga readers in the Bay of Plenty’s professional services, horticulture management, and logistics sectors, the practical takeaway is that the highest-exposure roles are also the ones with the highest augmentation pay-off: the same 10 LLMs the RBNZ scored say the judgement layer of a middle-band NZ job is the part that doesn’t commoditise.
Sources
- Stuff — The Kiwi jobs at highest risk of AI automation (MBIE briefing, released under the OIA)
- Reserve Bank of New Zealand — Assessing AI and Robotics Exposure in the NZ Labour Market Using LLMs, February 2026
- PwC New Zealand — New Zealand’s AI moment: the real opportunity is reinvention, not replacement, June 2026
- Related: NZ Employers Want AI Skills More Than Australian Ones Do · AI Skills Now Worth More Than an MBA, Say 86% of Finance Executives · The AI Jobs Apocalypse Never Showed Up, But Work Has Changed Anyway
