For the past two years, a certain kind of corporate announcement has become almost routine. The company cuts staff. The company cites AI. The company moves on. The formula was clean, the earnings calls were tidy, and the narrative was simple enough to fit in a headline. The only problem is that it wasn’t true, and the consequences are now showing up in ways that are hard to ignore. The companies that replaced workers with AI are discovering that the math doesn’t hold. According to CNBC, employers who laid off workers in the name of AI are reversing those decisions. About half of all companies that swapped people for AI end up experiencing a boomerang effect, rehiring at greater expense than it would have cost to retain the original workforce. The Great AI Layoff, it turns out, is turning into the Great AI Rehire.
Klarna became the poster child for AI displacement. The fintech company announced its AI chatbot was doing the work of 700 customer service employees and would contribute $40 million in annual profit. The coverage was everywhere. What got less coverage was what happened next. Customer satisfaction deteriorated, frustrated users took their complaints public, and Klarna began rehiring human agents. The chatbot could handle volume. It couldn’t handle nuance, de-escalate a genuinely upset customer, or exercise the kind of judgment that turns a bad experience into a loyal one.
Klarna isn’t alone. Analysis from Bloomberg suggests that job losses in Britain attributed to AI were actually driven by broader economic factors, meaning companies used the AI narrative as cover for cuts they were planning anyway. The result is a distorted picture of what AI does to employment, and a workforce that absorbed the damage of a story that wasn’t accurate.
I’ve spent 15 years researching the future of work. I’ve watched companies make this mistake before with automation, offshoring, and digital transformation. The pattern is consistent: Cut fast, overpromise the technology, underestimate the human capability you just removed, and spend the next several years trying to recover it.
The more instructive story isn’t the company reversing course. It’s the ones that never made the mistake in the first place. When Ingka Group, which controls the bulk of Ikea stores, trained an AI chatbot to handle 47% of its customer service calls, it faced a genuine choice. It could have laid off 8,500 workers. Instead, it retrained them as interior design consultants, investing in the human capabilities that AI couldn’t replicate. The result was €1.3 billion in revenue in 2024, a figure projected to reach 10% of total revenue by 2028. The workers weren’t a cost to be cut. They were the asset that made AI useful.
IBM reached a similar conclusion via a different path. After aggressive AI adoption, the company discovered that cutting entry-level hiring created a talent drought three to five years down the road. IBM is now tripling its Gen Z entry-level hiring, rewriting what those roles look like in the AI era rather than eliminating them. The jobs changed. The people didn’t go away.
Amazon Web Services is making the same bet. CEO Matt Garman is hiring 11,000 interns and recent graduates this year and says AWS employs more software developers today than it did two years ago, even as AI coding tools have become dramatically more capable. The tools made developers more productive. They didn’t make developers unnecessary.
Jeff Bezos made an argument at VivaTech in Paris that I think will age well. He predicted AI won’t eliminate jobs but create a labor shortage, because humanity has an essentially unlimited capacity to invent things that need to be built, and someone always needs to build them. The history of technological revolutions supports this view. Automation in manufacturing didn’t end employment. It shifted where employment happened and raised the skill floor for the jobs that remained.
The data is starting to reflect this. PwC’s 2026 Global AI Jobs Barometer, which analyzed more than 1 billion job postings across six continents, found that the companies making the most effective use of AI are pulling ahead in both productivity and hiring. The top 20% of AI-exposed companies achieved labor productivity growth of 163% relative to 2018, nearly five times the average, and recorded growth in head count of 52% versus 36% among the least AI-exposed firms. Using AI well doesn’t shrink your workforce. It grows it.
That said, the entry-level picture is genuinely complicated. A GMAC recruiters survey found that Gen Z entry-level roles in tech and manufacturing face real displacement risk. The answer isn’t to pretend that threat doesn’t exist. It’s to do what IBM and Ikea did: redesign the roles rather than eliminate the people.
The lesson from the companies getting this right is straightforward, even if it’s harder to execute than a layoff announcement. AI works best when it amplifies what people already do well. The organizations that treated their workers as the asset, not the obstacle, built something more valuable than efficiency. They built resilience, loyalty, and the kind of institutional knowledge that a chatbot cannot inherit.
The boomerang is real, and it’s expensive. Half the companies that cut workers for AI are now paying more to get them back. The smarter play, and the evidence is accumulating fast, was never to lose them in the first place.