AI Didn't Come for the Jobs. It Came for the Ladder.

This week, Stanford's SIEPR Economic Summit released the data everyone was expecting but no one wanted to see.

Entry-level software developer hiring is down 20% since ChatGPT launched. Call center jobs are down 15%. Early career workers in AI-exposed roles have seen a 13% relative decline in employment overall.

The same week, Atlassian cut 1,600 employees — 10% of its workforce — to "self-fund AI investments." Meta is reportedly cutting up to 16,000 people while committing $600 billion to AI infrastructure through 2028.

These aren't layoffs. They're architecture decisions.

The Middle Is Holding. The Bottom Isn't.

Here's what most of the coverage misses: the senior roles aren't disappearing. Erik Brynjolfsson's research at Stanford found that older, more experienced developers kept getting hired. It's the 22-to-25-year-old developers — the ones fresh out of school, the ones who were supposed to become your senior engineers in five years — who are being replaced.

This isn't about AI taking jobs from the top. It's about AI removing the bottom rungs of the ladder that people climb to get there.

And that's a fundamentally different problem than the one most leaders are preparing for.

Efficiency Today, Brittleness Tomorrow

When you cut entry-level roles to fund AI, you're making a trade that looks great on a quarterly earnings call. Faster output. Lower headcount. Cleaner margins.

But you're also making a bet that you'll never need those people to grow into the leaders who run your company ten years from now. You're betting that the institutional knowledge, the judgment, the relationships built over years of doing the work — that all of that can be replaced by a model.

Maybe it can. But probably not yet. And definitely not automatically.

The companies cutting their pipelines today are building organizations with no succession plan. They're optimizing for the current quarter at the expense of the next decade.

What Leaders Should Actually Be Doing

If you're running a technology organization right now, the question isn't whether to use AI. That's settled. The question is whether you're using AI to replace your talent pipeline or to accelerate it.

Those are two very different strategies with very different outcomes.

The leaders who get this right will be the ones who redesign entry-level roles around AI — not the ones who eliminate them. Junior developers who learn to work with AI agents from day one will become senior engineers faster, not slower. Call center staff who are trained on AI-assisted workflows will handle more complex customer relationships, not fewer.

The ladder doesn't have to disappear. But it does have to be rebuilt.

The Real Risk

The Stanford data tells us something uncomfortable: the market is already making this decision, and most organizations are letting it happen passively. They're not choosing to eliminate the pipeline — they're just not hiring into it, and calling it efficiency.

That's not strategy. That's drift.

The organizations that will lead in five years are the ones making deliberate decisions about talent development right now — not the ones celebrating how many headcount they eliminated this quarter.

AI didn't come for the jobs. It came for the ladder. And if you're not rebuilding it, you're just standing at the top of a structure that's quietly disappearing beneath you.