Sarah Choudhary is CEO of ICE Innovations and executive advisor with expertise on quantum AI, ethical technology and sustainable innovation.
A CTO at a mid-size SaaS company told me last spring that he had cut his QA team by 60% and let the AI handle it. He sounded proud. Six months later his team was rehiring, shipping slower and explaining to the board why three preventable production incidents had cost them an enterprise contract. He is not unusual. He is the pattern.
We keep talking about vibe coding and AI adoption as if the only question is whether developers will be replaced. That framing misses the story. The story is that a specific kind of executive, the one who needs a progressive headline every quarter, has been running an uncontrolled experiment on your workforce. The data on that experiment is now in, and it is not flattering.
Start with the reversal numbers. Gartner projects that 50% of companies that attributed headcount cuts to AI will rehire for similar functions by 2027. Forrester found that over half of companies that cut staff for AI already regret the move. A Careerminds survey went further. One in three employers spent more on restaffing than they saved from the original layoffs. That is not efficiency. That is a wire transfer with extra steps.
The underlying AI investment is not rescuing the math either. An IBM survey of 2,000 CEOs found only one in four AI projects produces the promised return, and just 16% ever scale across the enterprise. MIT researchers found that only 5% of companies fully embracing AI saw measurable profit from the investment. So the remaining 95% are cutting people to fund infrastructure that has not paid for itself.
Klarna made this pattern impossible to ignore. The Swedish fintech cut its workforce from 5,500 to 3,400 and publicly claimed its AI chatbot was doing the work of 700 customer service agents. Within a year, customer satisfaction fell on complex interactions. The company began rehiring agents in 2025, with CEO Sebastian Siemiatkowski admitting the firm had “focused too much on efficiency and cost.” Translation. We fired the people who understood our customers, then we had to rent that understanding back.
Consider the range of roles affected. This is not only about engineers. QA analysts, data analysts, copywriters, content moderators, support staff and software architects have all been caught in the same loop. IBM, Salesforce, Google, and Meta have all quietly restaffed roles they previously eliminated. The Challenger, Gray and Christmas tracker logged nearly 55,000 AI-attributed layoffs in 2025. Visier’s analytics team reported the highest employee rehire rate since 2018. And in a detail that should embarrass everyone involved, investigations revealed that Amazon’s “Just Walk Out” cashierless stores were partially run by remote workers in India reviewing video feeds, not the pure automation the marketing promised.
Here is what makes this worse than a normal bad tech bet. Most of these decisions were not made by people who understood the limits of the technology. Visier’s head of research, Andrea Derler, put the problem bluntly, noting that senior executives have not taken the time to understand what AI can and cannot do, according to PeopleMatters (linked above). IBM’s CEO survey (linked above) confirmed it. Sixty-four percent said the fear of falling behind drives them to invest in technologies before they understand the value.
Read that number again. Two out of three chief executives are buying a tool they cannot evaluate, cutting staff on capability claims they cannot verify and calling the result a strategy. On a conference stage, this is called vision. On a balance sheet, it is called an unfunded liability.
I am not anti-AI. I build and ship AI products every day. What I object to is the pattern where a CTO wins applause in Q1 for bold AI-driven headcount cuts, and then in Q4 is quietly rehiring those same roles through an offshore agency while attributing the reversal to “integration challenges.” That is not progress. That is a career arc dressed up as transformation. And it is costing shareholders, employees and customers at the same time.
So what should leaders do instead? Three things.
Measure the workflow before you touch the headcount. Run the AI against real production tasks for at least one full quarter. Track defect rates, escalation volume, customer satisfaction and security findings next to the cost savings. If the quality numbers drop, the savings are fiction.
Budget the reversal. Any business case for AI-driven workforce change must include the rehiring, retraining and reputational cost of unwinding the decision. If the math still works with reversal priced in, proceed. If it does not, you just avoided the next Klarna.
Demand specificity before automation. The rule my team uses is simple. If you cannot describe what the human did, with examples, you do not understand the job well enough to automate it. Generalities about “handling customer queries” or “writing boilerplate” are not descriptions. They are the fog that hides the part the AI cannot do.
The uncomfortable part of all this is that it is not a technology problem. It is a leadership problem. AI does not make bad executives worse. It gives them a faster way to prove they are bad. The leaders still standing in 2030 will be the ones honest enough to put the rehiring cost in the business case before the ink dries on the layoff letter.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
