A Generation Chose Their Degrees for a Labour Market That No Longer Exists

The ten fields Forbes identifies as losing ground to AI are among the most popular degree choices in universities across the world — and the students currently enrolled in them have a question to face

Sign up for my Substack daily AI newsletter here.

See my AI Training course portfolio for corporate Business Leaders here.

Article link: https://open.substack.com/pub/johanosteyn/p/a-generation-chose-their-degrees?r=73gqa&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true

Follow me on LinkedIn: https://www.linkedin.com/in/johanosteyn/

Somewhere right now, a first-year accounting student is sitting in a lecture theatre being taught double-entry bookkeeping. A communications graduate is completing a press release module. A paralegal student is learning to summarise case law. A graphic design cohort is practising logo construction in software their lecturer has used for fifteen years. Each of them has made a rational choice, based on available information, about which qualification would give them the best chance of a stable professional career. The information they based that choice on is out of date. The roles they are training for are being automated. And almost nobody in the institutions they are attending has told them.

Forbes contributor Jodie Cook published a list in April 2026 of ten college degrees AI is making redundant right now. The list includes generic business administration, basic marketing, journalism and media studies, communications, paralegal and pre-law tracks, basic computer science, accounting, finance, graphic design, and English literature in its purely academic form. These are not obscure or peripheral qualifications. They are among the most commonly chosen degrees at universities across the world. The students enrolling in them are not making poor decisions. They are making decisions based on a labour market that is changing faster than the advice they are receiving.

CONTEXT AND BACKGROUND

The data supporting the Forbes assessment is substantial and converging. EDsmart’s 2026 AI Job Automation Risk Report found that routine, rule-based, data-heavy office and administrative roles dominate the fifty occupations most at risk of automation, with an average of 86.3% of tasks automatable across those roles. Accounting specifically sits at 79.5% automatable tasks. Basic financial reporting, invoice processing, data entry, and reconciliation work — the tasks that constitute a junior accountant’s first years in practice — are precisely the functions AI tools are most directly replacing.

Goldman Sachs research has identified that younger tech workers are being disproportionately affected by AI-driven hiring headwinds, with unemployment among 20 to 30-year-olds in tech-exposed occupations rising by almost three percentage points since the start of 2025, notably higher than for their same-aged counterparts in other sectors. Goldman Sachs researchers note that this pattern is probably related to AI automation, and that it corroborates anecdotal reports that generative AI is contributing to hiring headwinds facing recent college graduates in technology specifically.

The Federal Reserve Bank of New York placed the underemployment rate for recent college graduates at 42.5% at the end of 2025, the highest level since 2020. ServiceNow CEO Bill McDermott has warned publicly that AI agents could send college graduate unemployment above 30%. BlackRock CEO Larry Fink has described the class of 2026 as potentially facing the highest graduate unemployment in years, even without a recession. These are not fringe predictions. They are assessments from the leaders of institutions with the most direct visibility of where hiring is going.

INSIGHT AND ANALYSIS

The Forbes list is not a verdict on the fields themselves. It is a verdict on how those fields are currently being taught. The distinction matters enormously, because the response it requires is not to stop studying accounting or journalism or law. It is to stop teaching those disciplines as collections of tasks that AI can already perform, and to start teaching them as disciplines that require the judgment, contextual understanding, ethical reasoning, and human relationship that AI cannot replicate.

The future accountant who will thrive is not the one who can reconcile a spreadsheet. It is the one who understands what the reconciled spreadsheet means for a business decision under uncertainty, who can identify the anomaly the algorithm flagged but cannot interpret, who can advise a client on risk in a way that accounts for the client’s specific situation and the adviser’s professional accountability for that advice. The future journalist who will thrive is not the one who can write a press release. It is the one who can investigate what the press release is concealing, build relationships with sources who will not speak to an algorithm, and make editorial judgments about what a community needs to know. Those capabilities are not taught in most curricula currently. They are not assessed by most examinations. They are not what most degree programmes are designed to develop.

I have previously written about the structural shift in what employment means in the age of AI, and the accountability question that arises when organisations deploy AI to replace human roles without acknowledging the full cost of that replacement. The degree crisis sits inside the same argument, one step earlier in the pipeline. When universities produce graduates whose primary skill set consists of tasks AI can perform, and when employers discover this on the first day of work, the cost of the mismatch falls on the graduate. The institution has moved on to the next cohort. The employer has begun to recalibrate its hiring model. The graduate is left holding a credential that certified something the market no longer values, having spent three or four years and significant financial resources acquiring it.

IMPLICATIONS

The curriculum lag is not a new problem, but AI has compressed the timeline in which it becomes critical. Universities typically review and update their programmes on cycles of five to seven years. The capabilities AI is automating are being automated on cycles of months. The accounting curriculum being taught today was largely designed before generative AI could produce financial analysis. The marketing curriculum being taught today was largely designed before AI could generate campaign copy, segment audiences, and optimise targeting simultaneously. The gap between what is being taught and what the labour market needs is widening with every semester that passes without a fundamental rethinking of what those degrees are actually for.

The institutions best positioned to close that gap are not necessarily the most prestigious ones. They are the ones willing to redesign assessment around judgment rather than task performance, to build industry partnerships that reflect the labour market of 2026 rather than 2016, and to be honest with prospective students about which capabilities their programmes are designed to develop and which ones they are not. That honesty is currently rare. Most degree programme marketing still describes graduate outcomes in terms of the roles they lead to, without acknowledging that those roles are being structurally transformed by the same technology the students are learning to use as a tool rather than as a competitor.

For employers, the implications are direct. Organisations that continue to recruit graduates on the basis of degree credentials without interrogating what those credentials actually certify are building their talent pipelines on a foundation that is quietly shifting. The graduate with an accounting degree who cannot do what the AI cannot do is not an accounting resource. They are a training liability. The organisations that will build durable talent pipelines in the next five years are those that define clearly what human capability they are actually hiring for, and work backwards from that definition to identify which educational pathways are producing it.

CLOSING TAKEAWAY

The degrees on the Forbes list are not beyond saving. The fields they represent are not disappearing. Accounting, journalism, law, marketing, and design will all exist in forms that require human beings for decades to come. What is disappearing, with considerable speed, is the version of those degrees that teaches students to perform the routine functions of the field rather than to exercise the judgment that makes the field worth having in the first place.

The students currently enrolled in those programmes deserve to know the difference. The advisers guiding prospective students deserve to understand it. And the institutions responsible for producing the graduates that economies and organisations depend on have an obligation to close the gap between the curriculum they are delivering and the labour market their students are about to enter. That gap is not a future concern. For the graduating class of 2026, it is already the present reality.

Johan Steyn is a prominent AI thought leader, speaker, and author with a deep understanding of artificial intelligence’s impact on business and society. He is passionate about ethical AI development and its role in shaping a better future. Find out more about Johan’s work at https://www.aiforbusiness.net