The entry-level job market is shrinking before our eyes. If you’re a graduate looking for that first "real" role, the ground is shifting. Former UK Prime Minister Rishi Sunak recently pointed out that artificial intelligence is "flattening" the jobs market for young people. He’s not just being a doomer. He’s highlighting a structural change in how companies hire. For decades, the ladder was simple. You get a degree, you take a junior role doing the "grunt work," and you learn the ropes. Now, AI does the grunt work. It does it faster, cheaper, and it doesn't need a coffee break. This isn't some distant threat. It's happening in law firms, coding shops, and marketing agencies right now.
Why the junior ladder is losing its rungs
We used to call it "paying your dues." In professional services, that meant spending your first two years summarizing documents, cleaning up spreadsheets, or writing basic press releases. Managers accepted that juniors weren't very productive yet. The investment made sense because those juniors became the senior managers of tomorrow. AI has broken that bargain. Recently making news in this space: The Eight Shadows and the President’s Word.
When a Large Language Model can summarize a 50-page legal brief in six seconds, why hire a paralegal? When a junior dev spent three days writing boilerplate code that a tool can now spit out in a heartbeat, the business case for that hire vanishes. Sunak’s point about "flattening" refers to this middle layer. The gap between "unskilled" and "expert" is becoming a cliff. Companies still want the experts, but they're cutting the path people use to get there.
The data backs this up. Recent reports from the Institute of Student Employers show that while overall hiring remains steady in some sectors, the nature of the work is changing. Firms are looking for "AI-native" graduates. They don't just want you to know how to use the software. They want you to prove you can do the work of three people by using these tools effectively. If you can't, you're too expensive. Further information into this topic are detailed by The Guardian.
The experience trap is getting worse
You've seen the memes. An entry-level job posting that requires five years of experience. It’s a joke, but it’s also a symptom. AI allows small teams to stay small. A startup that might have needed ten people five years ago can now run with three experts and a suite of automation tools. This means the "safe" corporate training ground is disappearing.
Young people are stuck. You need experience to get the job, but the roles that used to give you that experience are being automated. Sunak mentioned this during his discussions on the future of work, suggesting that the very concept of a "career path" is being rewritten. We're moving toward a world where you aren't hired for your potential to learn, but for your ability to deliver immediate value using technology.
Coding is no longer the golden ticket
For years, the advice was "learn to code." It was the ultimate career insurance. That insurance policy just got canceled. Tools like GitHub Copilot have turned coding into a commodity at the junior level. If an AI can write the syntax, the human just needs to understand the logic. This means the bar for "entry-level" in tech has skyrocketed. You aren't competing with other grads anymore. You're competing with a senior dev who is now 40% more productive because of AI.
The soft skills paradox
Here is the weird part. As technical tasks get automated, the "human" stuff becomes more valuable, but harder to prove on a CV. Empathy, negotiation, and complex problem-solving are things AI still struggles with. But how does a 21-year-old prove they have "executive presence" or "strategic intuition" when they've never been in a boardroom?
Sunak’s warning isn't just about jobs disappearing. It’s about the loss of mentorship. If you aren't in the office doing the boring work alongside a senior, you aren't absorbing the culture or the nuances of the trade. The "flattening" isn't just about salary or headcount. It's about the erosion of the apprenticeship model that has sustained professional life for centuries.
Real talk on what you should do right now
If you’re a student or a recent grad, panicking won't help. The "wait and see" approach is even worse. You have to change how you present yourself to the market. Stop listing "Microsoft Word" or "Basic Python" on your resume. That’s like listing "can breathe" as a skill.
- Build a portfolio of "AI-plus" projects. Don't just show that you can write a blog post. Show how you used AI to research, draft, and optimize ten posts in the time it takes someone else to do one.
- Focus on the "Edge Cases." AI is great at the average. It’s terrible at the weird, the niche, and the highly specific. Become an expert in a sub-sector that requires deep, messy, human context.
- Master the "Prompt." This isn't just about "Prompt Engineering" as a job title—which is mostly hype anyway. It’s about being the person who knows how to bend the tool to get a high-quality result.
- Network like it’s 1995. Since the digital front door is being guarded by AI filters and shrinking quotas, the "who you know" factor is back with a vengeance. Go to events. Send the cold emails. Get a human to vouch for you.
The government’s role is lagging behind
Sunak’s comments show a level of honesty we don't always get from politicians. Usually, they promise that "new jobs will be created" to replace the old ones. While that might be true in the long run, it doesn't help the person graduating this summer. The education system is still teaching students for a world that died in 2022. Universities are still grading essays that AI can write in seconds.
We need a radical shift in how we think about "skills." If the jobs market is flattening, the education system needs to broaden. We need more emphasis on "meta-skills"—learning how to learn, how to pivot, and how to manage technology rather than just being a cog in it.
The flattening is real. The ladder is broken. But for the people who figure out how to stand on top of the tools rather than underneath them, there is still a way up. You just have to stop looking for the old rungs. They aren't there anymore.
Start by auditing your own workflow. Pick a task you do every day and find a way to automate 80% of it using existing AI tools. Then, focus your energy on the 20% that only you can do. That 20% is your actual job description now. Get used to it.
