Will AI make your business degree worthless?

Analysis: The case against a business degree rests partly on the assumption that knowing how to use AI tools has made formal qualifications redundant

As the Leaving Cert nears completion and the CAO change-of-mind deadline approaches on 1 July, students who have spent two years working toward specific point targets may be feeling a sense of anxiety and second-guessing their choices. This year, there is more fuel for that anxiety than perhaps is usual.

Thousands of jobs have been cut at profitable technology companies in recent months, with AI cited as the cause. These announcements often get picked up on social media and turned into something broader: a general argument that business graduates are on the wrong side of what is coming and that a Commerce degree is no longer worth the investment.

Social media platforms are built to surface content that provokes the strongest possible reaction, which means a claim that your degree will be worthless will always travel further than a more considered assessment of what is actually happening to the graduate labour market.

We need your consent to load this rte-player contentWe use rte-player to manage extra content that can set cookies on your device and collect data about your activity. Please review their details and accept them to load the content.Manage Preferences

From RTÉ Radio 1’s The Business, how AI will change education

This is less a criticism of any individual creator than a description of the structural incentives that shape what gets amplified. A Leaving Cert student making one of the most consequential choices of their life deserves a more considered answer. Part of that argument rests on the idea that knowing how to use AI tools, ChatGPT for example, has made formal qualifications redundant.

What is a degree actually for?

The case against a business degree rests partly on the assumption that knowing how to use AI tools has made formal qualifications redundant. A working knowledge of ChatGPT is not the same thing as understanding what to do with its output. A language model can draft a market analysis in seconds. Working out whether the analysis is right, whether it applies to the organisation in question, and what should follow from it, requires a level of contextual judgement that technology cannot replicate. Developing that judgement is what university education is designed to do, and it is only part of what those years at college provide.

The classmates still in contact a decade on, the ones who put your name forward for something or know someone worth knowing, are what the sociologist Mark Granovetter called weak ties. 50 years of labour market research has found them consistently among the most significant factors in how careers develop. Universities build these over the years in ways that are difficult to replicate elsewhere, and graduate surveys confirm that personal connections remain the primary route into employment.

Developing that judgement is what university education is designed to do, and it is only part of what those years at college provide.

What AI is actually doing to work

This is not to suggest that routes into employment are as straightforward as they once were. The graduate jobs market is tough right now, but it is not simply technological change that is driving this. During the era of near-zero interest rates that ran roughly from 2010 to 2022, tech and financial services companies expanded headcount aggressively on the back of access to cheap capital. When that era ended, a correction was always coming.

What AI has provided is both the mechanism and the cover to make that correction more aggressively than the underlying economics alone would have demanded. Standard Chartered’s recent description of its own workers as “lower-value human capital” while announcing 7,000 redundancies is not careless language. It is a statement of values that treats employees as costs to be written down rather than people to be invested in. Meta’s decision to cut 8,000 jobs while posting record quarterly revenue of $56 billion makes the same point. These are strategic choices made by profitable companies, not the inevitable consequences of technological change.

The nature of the roles being cut is worth paying attention to. The Standard Chartered redundancies target back-office, compliance and operational functions, the kind of routine, rule-based work that has always been most exposed to automation. The World Economic Forum’s 2025 Future of Jobs Report finds the skills growing fastest in demand sitting alongside the technology shift, among them analytical thinking, leadership, social influence, and the capacity to work across organisations.

From RTÉ Radio 1’s Morning Ireland, Georgia-May Staunton from the National Youth Council of Ireland AI and Law Jury on research which shows young people are anxious about job displacement by AI

The MIT economist David Autor has spent decades documenting that the tasks hardest to automate require expert judgement, the management of relationships and sound decisions made under uncertainty. A business degree is designed to develop exactly those. The concern about technology making qualifications redundant is not new, and the historical record is worth reading carefully before drawing conclusions from the current narrative.

What the history of technology actually teaches us

The history of ATM is the most often-cited case study when it comes to understanding the technological impact of automation on work, but it is more instructive when read carefully than when used as a simple parable. When cash machines proliferated across the United States from the 1970s, economists expected bank teller employment to collapse. Instead, it nearly doubled over the following three decades, because lower branch operating costs made it viable to open far more branches. The teller role shifted toward relationship banking and financial guidance, and banks began recruiting more graduates into those positions precisely because the work had moved up the skill register.

The story has a second chapter that tends to be overlooked. From 2010 onwards, teller employment did eventually decline, but the cause was not the ATM. When mobile banking arrived and people could manage their finances from a phone, the branch visit itself became unnecessary. No machine had done the teller’s job, but the reason for the teller to be there had gone.

There is a reasonable argument that a well-chosen degree is among the more sensible places for an eighteen-year-old to be right now

To a large extent, this is what is happening at organisations like Standard Chartered and Meta. Compliance checking, routine data handling and back-office processing can now all be handled by automated systems at a fraction of the cost of employing people to do it, making entire categories of work surplus to requirements.

But business graduates do something different. The ability to make sound decisions in complex situations, build client relationships, and lead teams through difficult periods all depend on human judgment and the capacity to read people and organisations. These are the capabilities employers consistently say they need more of, and they are what a business degree is designed to develop.

From RTÉ Radio 1’s The Business, is it armageddon for Ireland’s tech industry?

The case for university

Most business programmes in Ireland have already incorporated AI literacy, data analytics and digital business into their courses. As the labour market works through its current correction and the real scale of AI investment becomes clearer, there is a reasonable argument that a well-chosen degree is among the more sensible places for an eighteen-year-old to be right now. The skills that will matter on the other side of this period are built in university, not picked up from a feed.

The attention economy has a particular appetite for content that generates anxiety, and the run-up to the CAO change of mind deadline are a reliable moment to exploit it. A business degree is not a guaranteed return, but it builds the capacity to understand organisations, work alongside people, and make judgements that require more than a well-formed prompt. A large language model cannot build those things for you.

Follow RTÉ Brainstorm on WhatsApp and Instagram for more stories and updates

The views expressed here are those of the author and do not represent or reflect the views of RTÉ

Tags: