AI and Entry-Level Jobs: The Hiring Pipeline Crisis

Something strange is happening at the bottom of the job market. Companies report intense demand for experienced, AI-literate talent, yet hiring for graduates and juniors has fallen sharply across the USA, the UK and much of Europe. The reason is uncomfortable but simple: the tasks juniors used to learn on, such as first-pass research, draft documents, routine code and support tickets, are exactly the tasks generative AI now handles well. The result is what we call the hiring pipeline crisis. The career ladder still exists, but its first rung is missing. Here is what the data actually shows, why it should worry employers at least as much as graduates, and what both sides should do next.

The Missing First Rung: What the Data Shows

The clearest evidence comes from Stanford's Digital Economy Lab. In a study titled "Canaries in the Coal Mine", Erik Brynjolfsson and colleagues analysed ADP payroll data covering millions of US workers and found that employment for 22-25 year olds in the most AI-exposed occupations declined 13% in the roughly three years since late 2022. The updated November 2025 version of the paper put the relative decline at 16% after controlling for firm-level shocks, with the effect still growing through 2025. Crucially, older workers in the same occupations held steady or improved. AI is not hollowing out whole professions yet. It is hollowing out the entry point.

The pattern repeats wherever you look. Employment of 22-25 year old software developers fell nearly 20% from its late-2022 peak by mid-2025, while workers over 30 in the same high-exposure categories actually grew 6-12%. SignalFire's State of Tech Talent report found new-graduate hiring by the 15 largest tech companies down more than 50% since 2019, with fresh graduates making up just 7% of new hires at big tech firms and under 6% at startups. The New York Fed puts unemployment for recent computer science graduates at 6.1%, well above the roughly 4.8% average for new graduates overall.

Nor is this an American quirk. UK graduate job openings fell 33% year over year in 2025 according to Indeed, the steepest decline in seven years, and Adzuna found UK entry-level vacancies down 32% since ChatGPT launched in late 2022. In Europe, Lufthansa plans to cut 4,000 administrative roles in Germany by 2030 as AI absorbs back-office work, precisely the kind of positions where graduates traditionally started. For the broader market picture, see our roundup of AI job displacement statistics for 2026.

Why Companies Stopped Hiring Juniors

Entry-level employment has always been a trade. The employer accepts lower output today in exchange for an experienced professional in three years. Generative AI broke the first half of that trade. Satya Nadella has said AI now writes 20 to 30% of the code on some Microsoft projects, and Sundar Pichai puts the figure above 30% for new code at Google. When a model produces the first draft, the first pass and the first triage, the early low-output phase of a junior's career starts to look like pure cost.

Boards have noticed. A Mercer survey of nearly 12,000 executives, HR leaders, investors and employees found the share of companies actively reducing junior and entry-level roles due to automation jumped from 17% to 43% in a single year. In coverage of the SignalFire report, 37% of managers said they would rather use AI than hire a Gen Z employee. In finance, a Citigroup report found 54% of financial-sector jobs have high automation potential, more than any other industry, and junior analyst recruitment has reportedly been cut by up to two-thirds at major banks. Even job descriptions tell the story: roughly 35% of positions labelled entry-level now demand years of prior experience.

Anthropic CEO Dario Amodei has warned that AI could eliminate up to half of entry-level white-collar jobs within five years. That forecast may prove too aggressive, but the direction matches what hiring data already shows, and software is the leading edge. We unpack that side of the story in Is AI Replacing Software Developers?.

The Pipeline Problem Nobody Is Pricing In

Here is the uncomfortable arithmetic: every senior professional was once a junior who built judgment by doing routine work under supervision. If an industry stops hiring 22-25 year olds in 2025 and 2026, it will be short of capable 28 year olds by 2029 and short of team leads by 2032. AI can write the first draft, but it cannot become your future head of engineering. Cutting the graduate intake is borrowing senior capacity from your own future, at a punishing interest rate.

The deeper irony is that research consistently shows juniors are the people AI helps most. A landmark study of 5,172 customer-support agents by Brynjolfsson, Li and Raymond found a generative AI assistant lifted issues resolved per hour by roughly 14-15%, with the largest gains going to the least experienced agents. A six-week GitHub Copilot trial at ANZ Bank in Australia found beginners improved 52% versus 40% for advanced users. In the Harvard and BCG study of 758 consultants, bottom-half performers gained 43% from GPT-4 against 17% for top performers. AI compresses the distance from novice to competent, which should make juniors cheaper to develop than ever, not pointless to hire.

Companies that over-cut are already paying for it. Klarna claimed its AI assistant did the work of roughly 700 customer-service agents, then reversed course and rehired humans after satisfaction dropped on complex cases, with CEO Sebastian Siemiatkowski admitting "We went too far." Careerminds found roughly two-thirds of companies that ran AI-led layoffs are now rehiring, and about one in three spent more on restaffing than the cuts saved. Forrester's 2026 Future of Work report estimated that 55% of employers regretted laying people off for AI-related reasons. Most telling of all: IBM, which replaced around 200 HR roles with AI agents, tripled its entry-level hiring for 2026, with its chief human resources officer noting that the work "still requires a human touch".

What Employers Should Do: Rebuild the Ladder, Apprenticeship-Style

The answer is not to recreate the junior role of 2019. That role really is gone. The answer is to redesign it the way skilled trades have run apprenticeships for centuries: small cohorts, structured progression, productive work from week one, and explicit investment in judgment rather than raw throughput.

In practice that means four moves. First, rebuild junior roles around AI supervision: the graduate's job becomes verifying, correcting and escalating AI output, owning quality on real work rather than producing volume. Second, make mentorship a budgeted commitment with named seniors and protected hours, because the routine tasks that used to create incidental teaching moments have disappeared. Third, hire smaller cohorts but never zero. A talent pipeline behaves like infrastructure: cheap to maintain, brutally expensive to rebuild. Fourth, pair the graduate intake with reskilling for existing staff, an approach we detail in our business guide to reskilling your workforce for AI.

The biggest employers are signalling the same conclusion. The World Economic Forum reports that 77% of companies plan to reskill or upskill workers between 2025 and 2030. Walmart is giving free AI training to 1.6 million US and Canada associates as part of a $1 billion skills investment while holding headcount steady. Microsoft's Elevate initiative commits $4 billion to credential 20 million people in AI, and Amazon's original upskilling pledge ultimately trained more than 700,000 employees, seven times its target. When we design business automation programmes at SpiderHunts, we push clients toward the same augmentation-first architecture: AI integration that absorbs the drudgery while deliberately preserving the supervised, judgment-building work that turns juniors into seniors.

What Graduates Should Do Right Now

If you are entering the market, the data points to a clear playbook. Get demonstrably good with AI tools, because the premium is real: PwC's 2025 Global AI Jobs Barometer, built on nearly one billion job ads, found roles requiring AI skills carry a 56% wage premium, and postings for AI-skilled jobs rose 7.5% even as total postings fell 11.3%. Lightcast's analysis of 1.3 billion postings found a 28% salary premium, worth nearly $18,000 a year. AI Engineer ranked as the fastest-growing job title on LinkedIn's 2026 list, with postings up 143% year over year.

Show shipped work, not just credentials. A graduate who can present three real projects built and verified with AI assistance is answering the exact question every hiring manager is now asking. Target augmentation-friendly roles: Harvard Business Review research found postings in the most automation-exposed roles fell 17% while demand for augmentation-friendly roles grew 22%. Finally, widen the aperture beyond Big Tech. Small and mid-sized businesses, healthcare, government and regulated industries across Canada, Australia and South Africa still run structured graduate intakes, and IMF analysis shows AI exposure varies widely between markets. For a personal strategy, see our guide on how to AI-proof your career.

The Bottom Line for Business Leaders

The long-run projections are not apocalyptic. The World Economic Forum expects 170 million new jobs and 92 million displaced by 2030, a net gain of 78 million. But net global figures will not staff your 2030 leadership team. The companies quietly keeping their graduate programmes alive, redesigned around AI, are buying the scarcest asset of the next decade: experienced people who grew up working with these systems. The cheapest senior hire of 2031 is a junior hired in 2026. We explore where this all settles in The Future of Work in 2030.

Frequently Asked Questions

Is AI really replacing entry-level jobs?

In exposed occupations, yes. A Stanford Digital Economy Lab study using ADP payroll data found employment for US workers aged 22-25 in the most AI-exposed occupations fell 13% since late 2022, and 16% after controlling for firm-level shocks, while older workers in the same occupations held steady or grew. The losses concentrate where AI automates work rather than augments it.

Why is graduate hiring down so sharply?

Employers are routing the routine work juniors used to do through AI. SignalFire found new-graduate hiring at the 15 largest tech companies fell more than 50% since 2019, UK graduate openings dropped 33% year over year in 2025 per Indeed, and a Mercer survey found the share of companies actively reducing junior roles due to automation jumped from 17% to 43% in a single year.

What is the hiring pipeline crisis?

Every senior professional was once a junior who learned by doing routine work under supervision. If companies stop hiring juniors because AI now does that routine work, the supply of experienced mid-level and senior talent dries up within five to ten years. The first rung of the career ladder is missing, and there is no second rung without it.

What should employers do about the missing first rung?

Redesign entry-level roles apprenticeship-style rather than deleting them: hire smaller cohorts, build the role around supervising and verifying AI output, fund deliberate mentorship, and set expectations that assume AI tooling from day one. IBM replaced roughly 200 HR roles with AI agents yet tripled entry-level hiring for 2026, and research shows juniors get the largest productivity gains from AI tools.

What should graduates do to get hired in 2026?

Build demonstrable AI skills. PwC's Global AI Jobs Barometer found jobs requiring AI skills carry a 56% wage premium, and AI-skill postings grew 7.5% while total postings fell 11.3%. Show shipped, AI-augmented work rather than credentials alone, target augmentation-friendly roles, and widen the search beyond Big Tech to SMBs, regulated industries and markets where structured graduate schemes still run.

Will entry-level jobs come back?

Partially, and in redesigned form. Forrester estimated 55% of employers regretted AI-related layoffs and predicts half will be reversed in some form by the end of 2026, and Careerminds found roughly two-thirds of companies that cut jobs for AI are already rehiring. The roles that return will be hybrid: humans supervising AI, with judgment valued over routine output.