Many young New Zealanders feel that a social contract has been breached. This tacit contract is that, if students worked and studied hard, the government would maintain through its institutions the macroeconomic stability and microeconomic flexibility that allow people to succeed with decisions on study and career choices.
However, young people are now expected to enter a weak labour market under tight fiscal constraints and with high uncertainty over the information learners can draw on. New Zealand has inherited multi-billion infrastructure deficits. It faces challenges in funding growing retirement and health liabilities. Its low domestic savings rates favour non-tradeable sectors and property rather than knowledge-intensive, export-orientated businesses.
Many tertiary institutions lack strong links with industry, and there is a lack of placements and internships for students. Some institutions impose “cultural competency” or Te Tiriti o Waitangi obligations on students – requirements that appear disconnected from New Zealand’s multi-culturalism and working life realities.
Looming over all is the fear that Artificial Intelligence (AI) will destroy jobs, polarise incomes, and create mayhem through identity fraud and online crime. Daron Acemoglu, a recent Nobel prize recipient argues that AI will destroy jobs and cause a massive rise in inequality.
However, such dire warnings about AI have not so far been vindicated and we need to take a closer look at the job market statistics. American statistics give us a richer picture than can be drawn from New Zealand sources.
Brynjolfsson et al. (2025) and other statistical sources suggest that AI may destroy tasks more so than the jobs themselves. AI has not yet had a detectably negative effect on the American job market, taken as a whole. US data shows that although the unemployment rate for the most AI-exposed workers is indeed rising, it is rising even faster for the least exposed workers. Unemployment rates have been rising for young workers and recent graduates alike, whether they are AI-exposed or not.
One pattern is clear in the data: highly exposed workers are doing better in the labour market than less exposed ones. Workers more exposed to AI are better paid, more likely to have bachelor’s or graduate degrees, and less likely to be unemployed than less qualified workers.
The US Census Bureau asks businesses whether they use AI in the production of their goods or services. A year and a half ago, only five percent said yes. The share is rising fast to about nine percent. There are some sectors where AI use is much higher than nine percent – about 27 percent of information businesses report they use AI.
The subsector with the highest use rate, ‘publishing industries’, where 36 percent of businesses report using AI, had a post-COVID surge in employment that then retreated and levelled off above the pre-COVID level, but is still below the 2021–23 trend.
Data processing and computing, which has the second highest AI use rate (35 percent), was rising until 2023 and has been flat since.
Since late 2022, employment has grown among most segments of the US workforce. Workers in their 30s, 40s, and 50s who are judged to be most heavily exposed to AI have seen robust employment growth since late 2022. Only very young workers who are also highly exposed to AI have seen their employment numbers fall.
One thesis is that older workers gain human management skills that complement AI, while younger workers are valued primarily for their narrower technical abilities.
Experience on the job is a complement to what AI can be taught in the classroom or lab. AI is likely to make on-the-job training a lot more important. This is good news if we can solve the problems associated with on-the-job training, which are “who pays for it?” how do employers choose the right recruit, and how can small businesses make the time to invest in the new recruit’s training.
To use current AI tools and capabilities a certain level of expertise and domain knowledge is needed. This means in practice that mid-career developers and managers are needed to set up AI systems in a way that makes sense. Entry-level workers on the other hand will almost uniformly lack the deep domain knowledge and company- or industry-specific expertise to set up AI correctly in a new job.
Key problems in our university system include its politization, its overemphasis on maximising student revenues, and the excessive focus given to purely academic achievement as opposed to wider outcomes such as making early impacts in a commercial environment.
Many students who complete the campus-based educational content of a qualification are unable to graduate because they have not found a company which can offer an internship that meets the student’s need for the degree’s work experience requirements.
In their 2025 report Amazing Grades: Grade Inflation at New Zealand Universities Dr James Kierstead with Dr Michael Johnston found the proportion of A grades had increased from 22% to 35% since 2006.
Pass rates are now mostly above 90% (and sometimes above 95%). These rises cannot be explained by better incoming students, more female students (who tend to get better grades), more funding, or a higher ratio of staff to students.
The grade inflation at our universities mainly reflects the incentive of earning more money by lifting student numbers, which means recruiting and retaining more students.
When most students receive top marks, employers cannot be certain that this reflects outstanding talent. High-achieving students see their accomplishments devalued. Grade inflation also reduces incentives on all students to work hard, as high grades seem guaranteed.
What are likely to be retained locally in New Zealand over the longer term are personal services ranging from gardening to surgery, from plumbing to old age care. Some such personal services will be low income because they require low skill levels or because they are difficult to automate and so capital and technology cannot be used to lift their productivity. Some of today’s high-earning professionals will become increasingly exposed. For example, much accounting, law and some parts of medicine can be offshored. These are effectively impersonal services in the sense that they can in principle be delivered from anywhere with an internet connection.
In the new world we are moving into, affinity with people, customer service aptitudes and social energy will be minimal threshold competencies rather than necessarily sources of competitive advantage. In many personal services a premium will be earned by those who can work with others, cross cultures, understand people’s innate psychological needs and act with wider interests in mind.
New Zealand is reforming its schooling system to strengthen science, technology, engineering and related fields. This will include enhanced links with industry, and a new emphasis on AI. This may potentially include a year 13 subject specifically on generative AI.
AI can seem universal with computational powers unmatched by our humble human brains. However, there are many people who blame social media and the ubiquity of mobile devices for all kinds of societal ills. These include hate speech, bullying, social isolation and loneliness.
People’s unique humanity gives us creativity and generative cultural products beyond a machine’s imagining. Only when AI can write like Dickens or compose like Mozart will it be able to pluck out the heart of our mysteries and then we will have a new world to contend with.
References
Brynjolfsson, E. Chandar, B. Canaries in the Coal Mine? 2025: Six Facts about the Recent Employment Effects of Artificial Intelligence. Stanford Digital Economy Lab. Stanford University Human-Centered Artificial Intelligence Working Paper.
Kierstead, J; Johnston, M. 2025: Amazing Grades: Grade Inflation at New Zealand Universities. NZ Initiative.
Peter Winsley 
