The Death of the Entry-Level Job: From Pyramid to Diamond

For generations, the path into a career followed a reassuringly simple logic: study hard, get a degree, land an entry-level job, and work your way up. Companies had to compete to draw the best fresh graduates, and if you were willing to carry out some pretty repetitive tasks to begin with, there was often a place for you at the bottom. However, it appears as if something is changing rather dramatically.

The pyramid that built the middle class

For most of the 20th century, companies were organised in a pyramid structure with a small number executives at the top, a broader level of managers in the middle, and a wide base of juniors at the bottom. These entry-level employees entered data into spreadsheets, scheduled meetings, proofread reports and kept the operation running. It was unglamorous work, but it was essential. For the young professionals, they learned how businesses functioned, and they picked up valuable new skills along the way.

Enter AI — and the disappearing base 

Unless you've been living under a rock, you'll know AI now drafts the emails, writes up the documents and crunches the data that used to land on a junior's desk. Which raises an uncomfortable question: what exactly is a graduate for and are they dispensable? In the US, graduates are now more likely to be unemployed than the workforce as a whole. Of those lucky enough to find employment, nearly 43% work in roles that never required a degree in the first place. Startups now only bring in graduates to fill 6% of roles, and even people with over 7 years' experience now fill up entry level roles once reserved for those straight out of university.

The disappearing middle 

With graduates slowly squeezed out, it seems inevitable that the pyramid is becoming a thing of the past. But what is happening to the managers in the middle? Managerial roles fell by over 6% between 2022 and 2025, and management now accounts for nearly a third of all layoffs, up from a fifth just a few years ago. 

That might suggest middle management is heading the same way as the entry-level, but the picture is less clear than the stats suggest. AI may generate options brilliantly, but often only within the framework that the human offers it. Left-field suggestions are often lacking, as AI still remains relatively keen to go along with whatever the human suggests. It also struggles to make the kind of judgment calls that actually run a business. So for now at least it seems as if human oversight is still indispensable, but no-one really knows for how long. 

If AI still proceeds only in whichever direction it is nudged, we will continue to require human judgment on AI decisions, meaning we may see something like a diamond: a thick section of middle managers, each running a very small, AI-augmented team of juniors. There may already be evidence for this, as one survey found over a third of employers had already rehired more than half the roles they had eliminated, most within six months. 

But if the layoffs in middle management keep going, and AI keeps closing that judgment gap, the diamond may turn into something closer to a tower, with executives, managers and juniors all compressed to a similar size. The technology is too new, and moving too fast, to say which one wins out. But for the graduate stuck at the bottom of the pecking order, the outlook is much the same: much fewer opportunities for entry.

The Dilemma for Businesses 

With everything we have covered, not hiring graduates is now seen as rational behaviour for many businesses. Training a junior employee is expensive because companies must spend money to train them before they see any ROI — and there is no guarantee that person won't leave after they are trained. So it now makes much more sense to hope that a competitor will take on a new graduate and invest in training them while you use AI to carry out all the entry level tasks. When you then need an experienced middle-manager, you can always take ready-trained talent from your competitors. 

Of course, if nobody wants to take on the burden of training a junior, who will only work slower than AI anyway, then there will eventually be no experienced middle managers left to hire. This could potentially lead to a decrease in productivity as opposed to the increase that we have been promised with AI.

Reasons for cautious optimism 

None of this is inevitable, and the picture is not entirely gloomy. The AI tools squeezing graduates out of entry-level jobs are also making it easier and cheaper to start something of their own. AI tools that would have required a team to operate five years ago are now accessible to a single motivated person, and the skills acquired could look very attractive to potential employers looking for middle managers. 

Instead of starting at the bottom, new graduates may need to embrace a more entrepreneurial mindset, start on their own, and then find a way in further up the pecking order as a middle manager.

Match each word to its definition and complete the example sentences in the exercise which can be found in the downloadable PDF.

According to the text, what were entry-level employees traditionally responsible for in the “pyramid” structure, and what did they gain from doing this work?

What evidence does the article give that middle management, not just entry-level roles, might also be shrinking?

According to the article, why does AI’s “well-documented flaw” mean that human oversight might still be indispensable?

What does the article suggest graduates could do instead of starting at the bottom of a traditional company hierarchy?

Do you think AI’s tendency to be “too eager to please” is a temporary limitation, or could it always be a barrier to fully automating judgment-based roles? Why?

If you were a new graduate today, would you rather compete for one of the shrinking entry-level jobs, or try to build something of your own? What would influence your decision?

I am a passionate linguist with over 14 years' experience with high level professionals.  In this lesson you will learn:

Want to work through this lesson with me?