In 2026, the most important story in higher education is not that the degree has vanished. It is that the degree has lost its monopoly. Research from Gallup and Lumina Foundation shows employers still value college credentials, yet only 54% say colleges are graduating students with the skills their organisations need, and 69% say recent graduates require a moderate or great deal of additional training to perform effectively. At the same time, roughly three in four employers say a college degree will be as important as, or more important than, it is now over the next five years. That is not a verdict of irrelevance. It is a warning that the degree still carries status, but no longer guarantees readiness.
That contradiction explains why the public mood has turned. In 2025, a poll found that Americans’ sense of the importance of college had fallen to a new low, and confidence in higher education remained far below where it stood a decade earlier. Yet current students remain strikingly confident that their degrees are teaching them career-relevant skills and will help them find work. The split is revealing. Students still experience higher education as formation and possibility; the wider public increasingly judges it as a high-cost product whose economic promises must now be proven, not presumed.
The reason this rupture feels sharper in 2026 is that World Economic Forum and LinkedIn are tracking the same underlying reality: the shelf life of skills is shrinking. The Forum’s Future of Jobs work says 39% of workers’ core skills are expected to change by 2030, while 63% of employers already describe skills gaps as a major barrier to business transformation.
LinkedIn’s latest skills data adds a cultural layer to that macro picture: AI literacy surged in 2025, AI business strategy is now rising quickly, and communication, adaptability and relationship skills are climbing alongside technical capabilities. In other words, AI is not simply changing what people need to know. It is changing how often they must update what they know, and how visibly they must prove it. That is why the labour market increasingly treats degrees as opening signals rather than finishing certificates.
Research published by Harvard Business Review found that generative AI is reshaping rather than erasing white-collar work: occupations most exposed to automation are seeing fewer openings and slimmer skill profiles, while augmentation-oriented roles are adding AI-related skills and rewarding workers who can use these tools intelligently. Meanwhile, the National Association of Colleges and Employers reports that employer hiring for the class of 2026 is essentially flat and that skills-based hiring is now used by 70% of employers in its survey.
The message is subtle but consequential: the market is no longer asking whether someone studied. It is asking whether they can translate learning into performance fast enough to matter.
The pressure is most visible at the entry level, where the old bridge between university and work is under strain. The Federal Reserve Bank of New York reported that labour market conditions worsened for recent college graduates at the end of 2025, with unemployment around 5.7% and underemployment climbing to 42.5%, its highest level since 2020. Recent reporting has captured the same reality from several angles: graduates still retain advantages over time, but the first rung of the ladder is wobblier, more contested and less forgiving than it was even a few years ago. That matters because early-career work has always done more than provide income. It has served as the training ground for judgment, pace, context and professional confidence. When that ground weakens, the degree looks less like a launchpad and more like an expensive waiting room.
AI has also altered what employers mean by “potential”. Recent reporting shows that even elite post-graduate pathways are being repriced around AI proficiency, while some employers are explicitly shifting from pedigree-led screening towards proof of technical fluency, experience and adaptable judgement. At the same time, demand for AI-specific proficiencies has more than doubled on major freelance platforms, even as employers continue to prize human strengths such as creativity, emotional intelligence and communication. The result is not a simple move from hard skills to soft skills or from degrees to certificates. It is a new hybrid standard: candidates are expected to be technically literate, behaviourally strong and visibly capable of learning in motion.
Into that gap has stepped the micro-credential. In Coursera’s own 2025 survey of students and employers, the value of these credentials was overwhelmingly affirmed:
The scale of the market helps explain the excitement. Credential Engine now counts more than 1.85 million unique credentials in the United States alone, offered by more than 134,000 providers. This is no side hustle at the margins of education. It is a fast-growing parallel economy of signalling, verification and career positioning.
But abundance is not the same as value. New work from the Brookings Institution found that relevance is everything: a worker’s first job-relevant non-degree credential is associated with a 3.8% wage premium, compared with 1.8% for a job-irrelevant one, and the gains are even larger for workers without bachelor’s degrees. That is promising. It is also a caution. The OECD has warned that the evidence base on micro-credentials remains limited overall, while UNESCO has argued that learners need much greater transparency about whether a micro-credential carries academic credit, labour-market recognition, or merely marketing gloss. The smarter reading of the boom, then, is not that small credentials replace big ones. It is that the market is searching for sharper, faster and more occupation-specific proof of capability.
That distinction matters inside hiring systems as much as it does for learners. The Burning Glass Institute argues that “credential fluency” — an employer’s ability to interpret non-degree credentials as meaningful signals of job-relevant skill — is now emerging as a competitive advantage. Its 2026 research found that employers in the top decile for hiring workers with non-degree credentials do so 11 percentage points more often than employers in the bottom decile when filling the same kinds of jobs. Yet the same institute has also warned that many public commitments to skills-based hiring have not yet translated into durable changes in actual hiring practice.
The implication is sobering: the market wants better signals, but it is still learning how to read them.
This is precisely why the “death of the degree” framing is too blunt to be useful. Degrees still offer what short credentials often cannot: breadth, intellectual formation, social capital, access to mentors, and in many sectors a recognised pathway into regulated or professional work. Gallup’s 2026 research shows that students and graduates still see strong career value in their degree experience, and employers still broadly prefer candidates with college qualifications even when those credentials are not always the formal minimum.
The reality is not that the degree has become worthless. It is that the degree is being demoted from a self-sufficient guarantee to one powerful signal among several.
That is the more honest diagnosis. The real shift is not from degrees to no degrees, but from static credentials to layered proof. A diploma may still get a candidate into the room. What increasingly determines whether they advance is whether they can demonstrate practical competence, work credibly with AI systems, and keep learning in public. The old model treated education as a one-time sorting mechanism. The model now emerging treats education as a lifelong signalling system in which formal qualifications, portfolios, work experience and targeted credentials all reinforce — or fail to reinforce — one another.
So what would better look like? Not a false choice between university and skills training, but a degree redesigned as the foundation of a broader evidence stack. That means:
The World Economic Forum reports that 81% of employers expect to prioritise experience when assessing skills. UNESCO is pushing for clearer recognition frameworks, Credential Engine is arguing for transparency that shows how credentials connect to one another and to wages, and NACE continues to find that employers value internships, hands-on experience and career-readiness skills even in a skills-first market.
The direction of travel is clear. Better education is becoming less about seat time and more about verified capability. The better shift, then, is from education as a sealed product to education as a verified, evolving relationship between knowing, doing and adapting.
Universities should stop defending the degree as though criticism of labour-market outcomes were an attack on learning itself. The stronger response is to make degrees more rigorous about outcomes and more honest about what else graduates now need: AI fluency, work exposure, durable human skills and credible evidence of performance. Employers, for their part, should stop oscillating between prestige hiring and vague complaints about skill gaps. If they want alternatives to degrees, they must build the systems to recognise them. The degree is not dead. But the old bargain — study first, signal once, and prosper for decades — clearly is.