The Missing Rung

Entry-level tech hiring has collapsed 73 percent year-over-year. U.S. entry-level job postings overall are down 35 percent since January 2023. Graduate unemployment hit 5.7 percent in the fourth quarter of 2025 — worse than at any point during the 2008 financial crisis. 

The bottom rung of the career ladder isn’t being disrupted. It is being removed. And the question nobody building these systems seems to be asking is deceptively simple: if you can no longer be a junior, how do you ever become a senior?

The experts can’t see it from there

The Stanford AI Index 2026 landed this month with a finding that should have been humbling: 73 percent of AI experts believe artificial intelligence will positively affect how people work. Just 23 percent of the public agrees. That is not a gap. That is two entirely different realities occupying the same sentence.

The tempting read is that the public is scared and the experts are right. The more honest read is that the experts are suffering from the oldest bias in the book: survivorship. The people most confident that AI is good for careers are the people who already have them. They have tenure, equity, established networks, and the kind of deep expertise that AI currently amplifies rather than replaces. Their view of the ladder is from the top. The bottom looks different when you’re standing where it used to be.

The Stanford report documents the divergence but doesn’t dwell on its cause. It notes that 64 percent of Americans believe AI will reduce jobs. It records that only 10 percent of Americans are more excited than concerned about AI in daily life. It flags that AI-documented incidents of harm rose from 233 to 362 in a single year. What it does not do is reconcile these facts with expert optimism. The gap is presented as a communication problem. It might be something deeper.

The generation fighting back

Gen Z is not merely anxious about AI. It is angry. According to a Gallup survey published in April, the share of 14-to-29-year-olds who describe themselves as excited about AI fell from 36 percent in 2025 to 22 percent in 2026. Anger rose from 22 percent to 31. Hopefulness dropped from 27 to 18. These are not soft sentiment shifts. This is a generation forming its identity against a technology that it increasingly views as being built at its expense.

The behavioral data is sharper still. Sixteen percent of currently enrolled college students have already changed their major because of AI. And 44 percent of Gen Z workers report actively sabotaging their employer’s AI rollout — not resisting, not dragging their feet, but deliberately undermining it. That is a remarkable number. Nearly half of the youngest cohort in the workforce is choosing subversion over adaptation.

As Fortune put it, Gen Z’s rejection of AI “isn’t irrational — it’s a verdict on everyone who failed them.” They watched the institutions that were supposed to prepare them for the workforce — universities, employers, government — fail to adapt in time. The anger is not about the technology. It is about the adults in the room who kept saying everything would be fine.

The five-year drought

The arithmetic of the missing rung is straightforward and has a time delay. Companies that cut junior hiring today will face a senior talent shortage in five years. The relationship is not subtle: juniors become mid-levels become seniors become the people who design systems, manage teams, and make the judgment calls that AI cannot. Remove the input, and the pipeline dries up. No amount of prompt engineering replaces a decade of accumulated professional judgment.

ServiceNow CEO Bill McDermott told CNBC in March that graduate unemployment could “easily go into the mid-30s” in the next couple of years. His company has already automated 90 percent of customer-service use cases that previously required humans. There is something bracing about an AI CEO quantifying the damage of the very technology he sells. McDermott deserves credit for the honesty. What he did not offer was a solution.

Some companies are experimenting with redesigned junior roles — positions built around supervising AI output rather than producing the work directly. Law firms are using AI to handle research so that young associates can spend more time with clients and mentors. Some tech firms are creating what amount to AI apprenticeships: structured environments where juniors learn by evaluating and correcting machine output rather than writing code from scratch. These are interesting experiments. They are not yet happening at scale.

Meanwhile, the global services industry — where hundreds of thousands of careers have been built from entry-level BPO and contact-center work upward into management, analytics, and operations — is watching the same question play out across emerging markets. If the first rung disappears in Manila and Bangalore as well as in Chicago and London, the talent pipeline problem becomes genuinely global. The few companies thinking seriously about this are redesigning roles, not eliminating them. That distinction may turn out to be the only one that matters.

What the ladder becomes

The career ladder was never just a staffing mechanism. It was a knowledge-transfer system. Juniors learned by doing the work that AI now handles — the repetitive drafting, the routine research, the grunt-level debugging. That work was tedious, but it was also where professionals built the pattern recognition and contextual instinct that made them useful at thirty-five. Remove the tedium, and you also remove the training ground.

The optimists in the Stanford report are not wrong that AI will create new categories of work. They are wrong to assume that the transition will happen cleanly, that the people displaced and the people hired will be interchangeable, and that the institutions charged with reskilling a generation will suddenly start performing at a level they have never previously demonstrated. Fifty-eight percent of 2024 and 2025 graduates are still looking for their first job. That is not a transition. That is a gap, and it is widening.

The question for every company deploying AI is not whether the technology works. It plainly does. The question is what happens to the pipeline of human expertise that every organization still depends on and will depend on for the foreseeable future. AI can do a junior’s tasks. It cannot yet do what a junior becomes. And if nobody is willing to invest in the becoming, the ladder doesn’t just lose a rung. It collapses entirely.

The question for your business

If you’re not hiring juniors today, who runs your company in 2031?

Read more thought leadership articles here:

• • From Sneakers to GPUs — AI’s New Player
  • The Bridge That Isn’t There