How AI could make uni degrees less valuable

March 22, 2026 — 4:45pm

You have reached your maximum number of saved items.

Remove items from your saved list to add more.

Save this article for later

Add articles to your saved list and come back to them anytime.

Getting a university degree has long been seen as a ticket to a decent-paying job, even if that link has been increasingly questioned in recent years.

But what if artificial intelligence fundamentally changes this equation, and degrees become less likely to secure white-collar work?

As companies everywhere rush to exploit AI, one of Labor’s top economic minds, Dr Andrew Leigh, explored this and some other big economic questions posed by the AI craze in a speech last week.

It left me thinking that even if the doomsday “jobpocalyse” scenarios of mass lay-offs don’t come true, we can be pretty sure that AI will have all sorts of other effects on the labour market, particularly for people in white-collar jobs.

AI-driven job cuts have become all too real lately, as tech giants such as Atlassian, WiseTech and Block have brutally swung the axe. More employers will inevitably follow suit, so it’s understandable that much of the AI discussion is about whether we’re heading for a dystopian future where the robots take over humans’ jobs, causing untold economic and social misery.

The week AI came for Australian jobs

Leigh, assistant minister for productivity competition, charities and treasury, is at the more optimistic end of this debate and believes the “jobpocalypse” scenario is unlikely, though the truth is that no one really knows.

But the main point he made was that AI is already affecting some key economic theories about the world of work and labour markets.

First, there’s much debate about which workers will be the winners and losers in an AI world.

Economists have long grappled with questions of how technology can affect people’s pay, and the distribution of income more generally. One popular theory that appeared to explain why some workers make more than others is the idea that big technological changes tend to help people who are more skilled, while hurting people who are less skilled.

For example, the rise of computers was good for the partner of the law firm because it allowed the business to run more efficiently, but it didn’t help people who used to work as typists. In this theory, there’s a “premium” paid to people who are more educated, skilled and able to use the new technology.

However, Leigh suggests this dynamic is less clear with AI. Instead, he says, there’s some evidence that AI can raise productivity disproportionately for lower-performing workers in the office, such as by helping people deliver better written reports.

If the machines start doing more of this sort of work, Leigh suggests the more important distinction in the future might be whether someone can show judgment, as opposed to technical skill.

“Rather [than] the formal skill categories, such as school completion, vocational qualifications and university degrees, the more relevant distinction in the future might be differences in the type of cognition performed,” he says.

“Potentially, the significant distinction might be judgment versus execution, oversight versus production, or conceptual reasoning versus procedural cognition.”

What might this mean in practical terms?

Well, it could mean people in “judgment-intensive” roles are in high demand, such as “the partner who signs the opinion, the surgeon who leads the procedure, the engineer who approves the plan, the judge who authors the decision”.

Leigh suggests it could also mean that employers look for “judgment” rather than technical skills – coding would be the most obvious example of a technical skill that’s been seriously threatened by AI. How does an employer test if someone’s got judgment? Perhaps through application tests, or by watching people on the job.

Another big economic idea that’s being challenged by AI is the idea of “human capital” – a theory that says people’s investment in education and training raise their productivity, thereby lifting future earnings.

How one CEO’s chatbot could cost his company $355 million

Measuring “human capital” precisely isn’t really possible, but economists have used proxies such as university degrees, and the grades people get at uni. However, AI systems that can help to write people’s essays for them have clearly changed that equation.

“Historically, educators could treat student output as evidence of accumulated knowledge. When AI mediates that output, the inference becomes less reliable,” Leigh says.

A possible consequence of this, once again, is that degrees may become a weaker signal to a prospective employer that a person has studied hard and developed their skills in thinking, analysing or writing.

Putting all this together, it’s difficult to avoid the conclusion that AI is going to make it harder for people to get a start in those entry-level jobs that allow you to get a start. After all, the best way to demonstrate sound judgment to an employer is probably by working for them and showing you’re competent.

I’d also add another risk that economists should be thinking about: will AI make the world a more unequal place by diverting more income to the owners of capital, as opposed to people-paid wages?

That certainly seems plausible. After all, AI really refers to high-tech machines owned by companies – just like machinery in factories. If AI can do some of the work done by people, businesses inevitably have the incentive to roll out these machines more widely and thereby increase their profits.

Importantly, none of this is certain, and history has shown that new technologies eventually create new jobs, as well as destroying some.

Another economist in Labor’s Treasury ministry, Assistant Treasurer Dr Daniel Mulino, last week painted a more optimistic picture. He said that that over the past 150 years, roughly two thirds of GDP has gone to labour and one third has gone to capital, and these shares have been “relatively stable”. Labour force participation had also increased over this period.

What’s more, he noted, this had generally happened through all manner of technological upheaval, from the mechanisation of agriculture to the IT revolution of the late 20th century.

Are bots coming for tech jobs, or is it ‘AI-washing’?

Colin KrugerSenior business reporter

Senior business reporter

Mulino was also careful not to make any predictions, but he did say these trends suggest that “with a bit of regulation and perhaps with a bit of prodding, we might imagine that we can see a plausible benign outcome on at least some fronts from AI, even if it is transformative”.

Ultimately, we are all in the dark about where precisely AI is taking us. But it seems inevitable that it will have major and sweeping effects for the labour market, and for the skills that employers value. Hang on for the ride.

The Business Briefing newsletter delivers major stories, exclusive coverage and expert opinion. Sign up to get it every weekday morning.

You have reached your maximum number of saved items.

Remove items from your saved list to add more.

© The Sydney Morning Herald