avalw news
Ethan BrooksEthan BrooksVIEW PROFILE →

The broken bottom rung: how AI is hollowing out the entry-level jobs that once launched careers

tech2026-08-22 · 3 min read · 0 reads

Graduate unemployment is climbing, entry-level hiring is cratering, and the tasks that used to train juniors are exactly what generative AI does best. But the story is more tangled than a simple 'robots took the jobs'.

For generations, the deal was simple: get a degree, land an entry-level job, and spend a few years doing the unglamorous work that turns a graduate into a professional. In 2026, that first rung of the ladder is quietly disappearing, and a growing body of data suggests artificial intelligence is a major reason why — though, as we will see, not the only one.

The numbers are sobering. According to the Federal Reserve Bank of New York, recent college graduates aged 22 to 27 hit a 5.7 percent unemployment rate in the first quarter of 2026, notably higher than the 4.3 percent national average. Even more striking, nearly 43 percent of new graduates are now underemployed, working in roles that do not require their degree — the highest rate since the pandemic.

A collapse concentrated at the entrance

What makes this downturn unusual is where it is concentrated. This is not a broad recession hitting everyone equally; it is a targeted erosion of the jobs at the very start of a career. Entry-level job postings across the United States are down roughly 35 percent since early 2023, and the squeeze is tightest exactly where automation is advancing fastest.

A working paper from Harvard University put a startling figure on it: at firms that have adopted generative AI, entry-level hiring has dropped by roughly 80 percent per quarter since 2023. That is not a gentle slowdown but a near-collapse, suggesting that when a company brings AI into its workflow, the first budget line it reconsiders is the junior hire.

Why AI hits the bottom first

The first rung of the career ladder — the junior job where you learn by doing routine work — is the one most exposed to automation.
The first rung of the career ladder — the junior job where you learn by doing routine work — is the one most exposed to automation.

The logic is uncomfortable but clear. Think about what a junior employee actually does in their first years in most white-collar jobs: debugging code, reviewing documents, drafting routine communications, cleaning and organizing data. These are intellectually real but largely repetitive tasks — and they are precisely the tasks that today's generative AI handles quickly and cheaply.

This has produced a phenomenon researchers call seniorization. Rather than hiring a novice and training them, employers now load junior job postings with senior-level expectations — judgment, stakeholder management, strategic thinking — because the routine work that used to fill a junior's day has been automated away. The ladder has not just lost its bottom rung; the second rung has been raised out of reach.

But is AI really the culprit?

Here honesty demands caution, because the picture is genuinely contested. A May 2026 study from the London School of Economics found that remote work is actually a better statistical predictor of the decline in entry-level hiring than AI is. Other analysts argue the real driver is simply a weak, cautious hiring environment across the economy, with AI serving as a convenient explanation rather than the root cause.

The likeliest truth is that several forces are compounding at once: a soft macroeconomic climate, the lingering restructuring of remote work, and the arrival of AI tools all landing on the same vulnerable category of jobs simultaneously. Untangling exactly how much blame belongs to the algorithm is, for now, more art than science — and anyone claiming certainty is overselling.

The long shadow over the career ladder

Whatever the precise cause, the long-term risk is the same and it is serious. If companies stop hiring and training juniors, they are quietly eating their own seed corn. Today's senior professionals became experts by doing years of the very entry-level work that is now being automated. Remove that apprenticeship, and the industry may face a shortage of experienced talent a decade from now that no AI can quickly replace.

For now, the graduates who are succeeding are those who treat the market as fundamentally changed. Openings still exist in healthcare, cybersecurity, skilled trades and AI-native graduate programs, and the people landing them rely far more on internships, portfolios and personal referrals than on cold applications. The message for the class of 2026 is stark but not hopeless: the bottom rung has moved, and the way up now demands proof of skill, not just a diploma. The harder question is whether employers, in optimizing away the junior job, are quietly dismantling the machine that produces the experts they will desperately need tomorrow.

Ethan Brooks
Stay updated
Ethan Brooks
Subscribe to get an email whenever Ethan Brooks publishes a new story. No spam, unsubscribe anytime.
Ethan Brooks
WRITTEN BY THE AUTHOR
Ethan Brooks
2026-08-22 · 3 min read · 0 reads
View profile →
VERIFY THIS STORY
ASK AI
MORE FROM Ethan Brooks
Report this articlesupport@avalw.com