The Great Swap: Tech Fires 51,000 Workers, Hires Them Back as Someone Else

Meta cut over 1,000 jobs from Reality Labs in January. Same week, it posted hundreds of openings for AI researchers.

Block fired half its workforce. CEO Jack Dorsey just announced they're hiring AI engineers.

The jobs aren't disappearing. They're being swapped. And the person losing their job isn't the person getting the new one.

The Numbers Nobody Connects

51,000 tech workers laid off in 2026 so far. That's 843 people per day.

Meanwhile, tech companies are offering AI researchers $200 million compensation packages. Not for executives. For engineers who can build the next generation of models.

The World Economic Forum says 92 million jobs will be displaced by 2030. But 170 million new ones will emerge. Sounds balanced. Until you realize: these aren't one-to-one swaps.

The marketing coordinator at Meta who got cut from Reality Labs can't apply for the AI researcher role that opened the same day. Different skills. Different education. Different universe.

What the Swap Looks Like

Meta's Reality Labs employed roughly 15,000 people working on the metaverse. That team got hit with a 10% cut — over 1,000 jobs. CTO Andrew Bosworth announced the layoffs in an internal post. The company's redirecting resources toward AI.

Translation: we're done betting on virtual reality. We're hiring AI people instead.

Block cut 40% of its workforce — roughly 5,000 people. CEO Jack Dorsey said it out loud: AI automation drove the decision. Now Block's hiring "senior AI talent."

The laid-off customer support worker and the new AI engineer will never meet. But the severance package for one funded the signing bonus for the other.

The Training Myth

Here's what economists say: workers can retrain.

Here's what the data says: no, they can't.

84% of workers cite lack of AI-related training as the biggest barrier to adapting. Brookings studied the US Job Training Partnership Act, which spent billions on worker retraining programs. The result? No statistically significant improvement in employment rates, earnings, or job stability.

Retraining works in PowerPoint slides. It doesn't work when you're 45, just lost your job at a tech company, and the new openings require a PhD in machine learning.

The gap between "AI will create more jobs than it destroys" and "the person losing their job can get one of those new jobs" is the difference between a policy paper and a life.

Who Gets the New Jobs?

A recent grad with a computer science degree and internship experience at OpenAI.

Someone who spent the last three years building transformer models.

A researcher poached from DeepMind with a $10 million signing bonus.

Not the 40-year-old who spent a decade managing customer support tickets.

The jobs being created aren't entry-level. They're highly specialized. CBS News reports that companies are pulling back on hiring because they've realized they can "do more with less." That means fewer total jobs. And the ones that exist require skills most displaced workers don't have.

Same Company, Different People

January 2026 saw the highest tech layoffs in 17 years. Same month, just 5,306 new hires were announced — the lowest January hiring since 2009.

The math doesn't work. Even if all the laid-off workers retrained perfectly, there aren't enough jobs for them.

Tech isn't killing jobs. It's swapping the people who fill them. The VR engineer becomes obsolete. The AI researcher becomes indispensable. Same company. Different human.

And the person who got swapped out? They're competing for roles that no longer exist in an economy that moved on without them.

What This Actually Means

"AI creates more jobs than it destroys" is technically true. It's also useless if you're not the one getting hired.

170 million new jobs by 2030 sounds great. Until you realize 92 million people lose theirs first. And most of those 92 million can't do the work the 170 million jobs require.

The great swap isn't about technology. It's about who gets left behind when the economy decides it needs different people.

Meta's 1,000 laid-off Reality Labs workers and Meta's 200 new AI researcher hires will never cross paths. But they're the same story.

One just funded the other's salary.