I recently read an article in the Sydney Morning Herald that discussed the issue of dwindling entry-level jobs, linked to the impact of AI on the workforce. It’s one of the first articles I’ve seen which discusses what I call the “pipeline problem, “ and I’m glad this is starting to come into the conversation.
Entry-level hiring is slowing while AI-related job postings climb. On the surface, it sounds like the classic labour market shift: old jobs disappear, new jobs emerge. Move along, nothing to see here. But there’s a problem hiding inside that narrative, and it’s one nobody’s quite saying out loud yet.
What the data shows
Entry-level workers aged 22–25 in AI-exposed fields like software development and customer support have seen employment drop by 16% since late 2022. Simultaneously, AI engineer postings are climbing, with 75,000 AI-related job openings for entry-level roles in the U.S. alone between 2023 and 2025.
On paper, it’s a wash. But in practice, it’s not. The jobs that are disappearing aren’t being replaced one-to-one. They’re being absorbed by AI. What’s being created instead are roles that demand skills most new graduates simply don’t have yet.
Why entry-level jobs matter beyond the paycheck
I’ve been thinking about what entry-level roles actually do, beyond the obvious: they’re not just a way to pay rent. They’re where judgment is built.
You do the grunt work—the data entry, the junior analysis, the customer research. You do it alongside someone who’s been doing it for ten years. You ask dumb questions. You make mistakes that cost nothing. You learn what works in practice, not just in theory. You develop judgment.
Judgment is the thing that can’t be automated. It’s the ability to look at AI output and know whether it’s plausible, or wrong, or right-enough-for-now. It’s the tacit knowledge that comes from having done the work yourself. Senior workers have “tacit knowledge, they learn tricks of the trade that maybe never get written down,” and those are the things AI hasn’t yet learned to replicate.
The risk we’re not talking about
If entry-level roles shrink, who gets to build that judgment? Entry-level job postings plunged 35% from January 2023 to June 2025, and while companies are cutting entry-level roles, researchers warn that “junior positions are shrinking at companies integrating AI” and that “AI is eroding the ‘bottom rungs’ of career ladders”. One expert put it starkly: “How in the world are young people going to get trained up to come in at a ‘Level Three’ if they haven’t done Level One and Level Two?”
We’re all talking about “human in the loop” as though it’s a given. We assume there will be humans capable of stepping in, of understanding nuance, of knowing when an AI recommendation doesn’t smell right. But that capability isn’t innate. It’s built through years of doing the work. If we’ve removed the space where that building happens, we’ve removed the training ground without building an alternative.
What some organisations are already trying
This isn’t a problem that organisations are ignoring. Many are genuinely grappling with it. Some are:
Redesigning junior roles to focus on higher-order thinking while AI handles the repetitive work
Experimenting with expanded internship programmes or rotational schemes that build depth even when full-time junior positions are scarce
Creating peer-learning structures where judgment is shared horizontally rather than top-down
Building capability assessments that help junior staff understand where AI outputs are reliable and where they’re not
But these are workarounds, not systemic solutions. And they’re not available everywhere.
What might actually shift this
If you’re leading an organisation or involved in hiring decisions, this is worth asking:
How are we building judgment in our junior staff if we’re automating the work that used to do that?
What happens to our ability to validate and govern AI if nobody in the next cohort of leaders ever did the foundational work?
How do we create apprenticeship pathways that don’t depend on traditional entry-level roles?
The conversation we should be having isn’t really about whether AI will displace young workers. It’s about what happens to expertise itself when the pathway to building it disappears. And what we’re prepared to do differently so we don’t lose a generation of people who never got the chance to develop judgment in the first place
.
Sources
Axios (August 2025). “AI is already taking jobs away from entry-level workers.” https://www.axios.com/2025/08/26/ai-entry-level-jobs Stanford ADP payroll data: 16% employment decline for ages 22–25 in AI-exposed roles since late 2022.
CNBC (November 2025). “Why AI may kill career advancement for many young workers.” https://www.cnbc.com/2025/11/20/why-ai-may-kill-career-advancement-for-many-young-workers.html Revelio Labs analysis showing 35% drop in entry-level job postings (January 2023–June 2025) and career ladder erosion.
CBS News (April 2026). “This is the fastest-growing job for young workers, LinkedIn says.” https://www.cbsnews.com/news/artificial-intelligence-entry-level-role-linkedin-study/ LinkedIn data: 639,000 AI-related U.S. postings (2023–2025); 75,000 specifically for AI engineer roles.
Jobs and Skills Australia (August 2025). Our Gen AI Transition – Implications for Work and Skills. https://www.jobsandskills.gov.au/studies/generative-artificial-intelligence-capacity-study Landmark government study of AI adoption across Australian industries; notes early evidence of recruitment shifts away from junior positions, particularly in technology sector. Emphasises that “while we haven’t yet seen an impact on entry-level roles in Australia, it will be important that the labour market continues to provide these valuable formative roles.”
Information Age / ACS (September 2025). “’We hire fewer entry-level engineers’: Aussie firms on AI impact.” https://ia.acs.org.au/article/2025/-we-hire-fewer-entry-level-engineers---aussie-firms-on-ai-impact.html Analysis of Jobs and Skills Australia’s Generative AI Capacity Study. Documents case studies from Australian companies (ReadyTech, medical research institutions, legal firms) where employers have shifted toward hiring senior staff and reduced junior intake.
CAQA Compliance (April 2026). “If Entry-Level Jobs Disappear, What Happens to VET Pathways?” https://caqa.com.au/blogs/news/if-entry-level-jobs-disappear-what-happens-to-vet-pathways-rethinking-work-based-learning-in-the-age-of-automation Analysis of automation trends in Australia; notes youth unemployment climbed to 10.5% in late 2025 (highest since 2021) and examines implications for vocational education and work-based learning pathways.
SMS Personnel (March 2026). “Gen Z Jobs Australia: Stand Out in the AI Era.” https://smspersonnel.com/gen-z-jobs-australia-ai-stand-out/ Reports findings from Jobs and Skills Australia and Foundation for Young Australians’ New Work Reality report showing 50% of 25-year-olds in Australia struggle to secure full-time work; discusses AI’s impact on frontline and entry-level roles, particularly affecting young workers and women.
itbrief (November 2025). “AI reshapes entry-level jobs & skills in Australian workplaces.” https://itbrief.com.au/story/ai-reshapes-entry-level-jobs-skills-in-australian-workplaces Survey data: 98% of Australian organisations report AI is impacting hiring and competencies; 30% have reduced junior staff intake, with 93% planning further decreases within five years.