AI is transforming the world of work. Is it coming for your job? A new study identifies 82 roles at risk of disruption.

Artificial intelligence is messing up everything. Pope Leo, for example, spitballed the other day about “an instrument of domination, exclusion, and death”.

So there’s that. But for many of us it’s a more banal and immediate question: is the proliferation of this technology – spanning machine learning, generative AI, robotics and automation – about to make us redundant?

New Zealanders seem to be more worried than most. A new Ipsos survey of 32 countries asked about the impact on the job market across the next three to five years. It found 58% of us thought the technology would make things worse, with just 14% saying nah fine. We’re the most pessimistic place in the world.

The findings are consistent with various research pointing to serious levels of AI distrust in New Zealand, said Professor Ali Knott, a specialist in cognitive science and artificial intelligence at Victoria University of Wellington. Any number of factors could contribute to that attitude, he said, but it could be informed by the signals from the top.

The current government had been “fairly evangelical” about AI without simultaneously setting out “the necessary transparency or regulatory mechanisms”, he said. Distrust may also stem from the areas that organisations using AI prioritise. Deloitte NZ’s Human Capital Trends Report, published in April, found that “trust, ethics and data governance risk becoming blind spots”.

New Zealand organisations are placing less emphasis than their international counterparts around the world on “AI ethics, workforce data trustworthiness, and managing AI’s human impacts – areas that become more critical as AI use scales”. Just 50% of New Zealand respondents, compared with 69% globally, rated balancing benefits, risks, ethics and potential conflict in AI as very or extremely important.

That was of particular concern, the report noted, given the unique importance of Māori data sovereignty.

The Ipsos survey was completed before the most recent headlines about AI replacing human work, as advanced by the finance minister, Nicola Willis. She urged public sector workers, who had for too long been “scared of AI”, to embrace the technology to offset the cut in head count demanded by last month’s budget. 

The public servants of Lambton Quay are not the only ones worried about the robots. A study published in Australia this month, for example, identified 82 different occupations that face serious disruption. More on that in a moment. 

What we know and what we don’t

In the US, artificial intelligence is now the main reason cited for cutting jobs. Some of that is real, some of it is “AI washing”, according to many, including Sam Altman. Perhaps the most visible figurehead in artificial intelligence today as the outspoken boss of OpenAI, Altman said in January, “there’s some AI washing where people are blaming AI for layoffs that they would otherwise do, and then there’s some real displacement by AI of different kinds of jobs.” 

Back in 2019, Altman had envisioned a scorched earth, saying AI would “probably replace most of the jobs people do today”. Speaking at a Sydney conference last month, however, he said he had overestimated the likely impact, and advances in AI would not lead to the “jobs apocalypse” he had earlier feared. 

Whether projecting apocalypse or merely a blizzard, every forecast in this territory needs to come with heavy caveats.

Knott was one of three New Zealand authors of a substantial 2021 study examining the ways that AI is set to impact work and employment. Back then, they were disinclined to make specific predictions of what was to come. After five years, and bearing in mind that Chat GPT was not released until 2022, Knott says he sticks by their approach, of encouraging those designing policy to look at different scenarios from a macro perspective.

“I much prefer the idea of saying we do not know yet, and what we should do instead is formulate different scenarios,” he said. “A lot of AI future tasks are so uncertain that they require fast reactions from government. I think all we can do is organise the possibilities into coherent alternatives, and then say: keep an eye out for this thing or that thing.”

The first was a positive vision in which “there are no widespread job losses, because productivity means that everyone can just do more, and everyone makes more money”. Another imagined “widespread job losses, but with some of the profits from AI remaining onshore, because the productivity, the automation comes from New Zealand-owned companies. If that’s the case, there are immediate options for redistribution of AI profits back to the community, so that you can address the challenge of people who’ve lost their jobs.” 

Then there’s the “bad scenario”, said Knott. That imagined “widespread job losses, and all the AI profits flowing offshore. That’s the really damaging one, because then you don’t have such an easy way of redistributing the money that’s gone.”

New Zealand could learn from instructive scenario-based work, said Knott, published by the OECD and the UK government. 

But what is AI doing to jobs, right now?

The bleak horizon

In a discussion paper for the Maxim Institute last year, Paul Henderson surveyed international research on AI and its reverberations. 

“The degree of job creation or loss caused by AI’s entrance into the workplace is contested. Some authors anticipate high unemployment within five years, while others are less convinced, viewing AI as an assistant to human labour rather than a replacement, with potential to increase national productivity.”

Inescapably, however, the overarching picture was “very bleak”, said Henderson, “with potentially 40-60% of secretarial, administrative and customer service roles exposed to and replaced by AI.”

The report by Maxim, a New Zealand public policy thinktank, included a table estimating various occupations’ exposure to the rise of AI. 

The clear conclusion, said Henderson, was the need “to prepare for AI’s rapid integration into work”. That preparation, he stressed, should include coming to terms with “statistical evidence [which] indicates that there is a need to recognise not only gender but also age and ethnicity as compounding factors on the impact of AI upon employment”.

Henderson pointed to sobering 2024 commentary by Kristalina Georgieva of the International Monetary Fund. She said: “In most scenarios, AI will likely worsen overall inequality, a troubling trend that policymakers must proactively address to prevent the technology from further stoking social tensions.”

As well as the roles that will be diminished or eliminated by AI, some will be enhanced or entirely new. Including some, said Knott, entrusted with keeping an eye on the technology and its deployments. “There will need to be a whole new economic sector,” he said.

“It’s going to take a lot of work to keep this stuff safe and make it work the way we want it to. That involves a mix of technical understanding and policy work, and then a bunch of social science to do with [assessing] algorithms … An enormous amount of work and effort is going to be needed. I think that AI regulation, AI governance, AI social impact measurement, AI policy making will become an important new area of work.”

Winners and losers in the AI age

Earlier this month, another Deloitte report generated many headlines across the Tasman. Heralding 2026 as “the year of AI”, it noted that there had been no widespread loss of jobs so far, suggesting that the technology, rather than eliminating occupations, is currently playing more of an “augmentative role”.

It identified several occupations likely to be “AI-enhanced”, in which they expect to see stronger demand.

• Chief Executives and Managing Directors
• General Managers
• Legislators
• Research and Development Managers
• Artistic Directors, Media Producers and Presenters
• Life Scientists
• Medical Laboratory Scientists
• Primary School Teachers
• Education Advisers and Reviewers
• Occupational and Environmental Health Professionals
• Nurse Manager

But it also produced a list of more than 80 roles in which “core tasks” could be automated by AI.

Belt in. Here’s the list.

Good luck out there.

• Corporate Services Managers
• Human Resource Managers
• Engineering Managers
• Production Managers
• Supply, Distribution and Procurement Managers
• Commissioned Officers (Management)
• Licensed Club Managers
• Other Accommodation and Hospitality Managers
• Retail Managers
• Amusement, Fitness and Sports Centre Managers
• Call, Contact Centre and Customer Service Managers
• Conference and Event Organisers
• Transport Services Managers
• Accountants
• ICT Trainers
• Training and Development Professionals
• Librarians
• Other Information and Organisation Professionals
• ICT Sales Professionals
• Graphic and Web Designers and Illustrators
• Other Engineering Professionals
• Medical Imaging Professionals
• ICT Business and Systems Analysts
• Multimedia Specialists and Web Developers
• Software and Applications Programmers
• Database and Systems Administrators, and ICT Security Specialists
• Computer Network Professionals
• ICT Support and Test Engineers
• Telecommunications Engineering Professionals
• Agricultural Technicians
• Civil Engineering Draftspersons and Technicians
• Electronic Engineering Draftspersons
• Mechanical Engineering Draftspersons
• ICT Support Technicians
• Graphic Pre-press Trades Workers
• Gallery
• Tourism and Travel Advisers
• Travel Attendants
• Contract
• Office Managers
• Practice Managers
• Personal Assistants
• Secretaries
• General Clerks
• Keyboard Operators
• Call or Contact Centre Workers
• Information Officers
• Receptionists
• Accounting Clerks
• Bookkeepers
• Insurance, Money Market and Statistical Clerks
• Betting Clerks
• Filing and Registry Clerks
• Mail Sorters
• Survey Interviewers
• Switchboard Operators
• Other Clerical and Office Support Workers
• Purchasing and Supply Logistics Clerks
• Transport and Despatch Clerks
• Conveyancers and Legal Executives
• Court and Legal Clerks
• Debt Collectors
• Human Resource Clerks
• Inspectors and Regulatory Officers
• Library Assistants
• Other Miscellaneous Clerical and Administrative Workers
• Insurance Agents
• Sales Representatives
• ICT Sales Assistants
• Motor Vehicle and Vehicle Parts Salespersons
• Pharmacy Sales Assistants
• Retail Supervisors
• Service Station Attendants
• Other Sales Assistants and Salespersons
• Checkout Operators and Office Cashiers
• Models and Sales Demonstrators
• Telemarketers
• Ticket Salespersons