AI Boomerang: 32% of Companies Rehired Employees Laid Off Because of AI

⚡ The Gist in 5 Seconds

• AP: Microsoft, Barclays, and Kroger cut staff citing AI — some of those laid off are now being brought back into the same or similar roles.
• A Robert Half survey (2,000 hiring managers in the US): 32% say their company eliminated a position because of AI and then hired a human back.
• Gartner: by 2029, 30% of employees laid off because of AI will have to be rehired, often at a noticeably higher cost.

🔍 What Happened

On October 29, 2026, AP published a piece by Katie Basevitz about a phenomenon that has already earned its own name — the "AI boomerang." Companies cite artificial intelligence when cutting staff: recent examples include Microsoft, Barclays, and the Kroger supermarket chain, with tens of thousands of people affected. Some of these positions are coming back: AI models, apps, and agents are performing worse than expected in practice, and people are being called back (Associated Press).

The mechanics of the return are more concrete than they might seem. Since last year, Klarna has been hiring back human support staff — CEO Sebastian Siemiatkowski admitted in a Bloomberg interview that the bet on an AI chatbot for the sake of savings had gone too far. In June, Ford announced it had hired or promoted 350 experienced engineers: its AI quality-control tools failed to deliver the expected level. IBM automated about 94% of routine HR requests, but the remaining 6% — including ethical dilemmas — required a human. The Robert Half survey, covered by Fast Company, breaks it down by industry: people were brought back most often in finance — 44%, in HR — 35%, in technology — 32%, followed by marketing, legal, healthcare, and administrative support.

🧠 Our Take

Skepticism is warranted here, but not in the direction it usually goes. The phrase "AI didn't replace people" sounds like an excuse, but it's actually about something else: companies did a poor job of mapping their own processes. Georgetown professor Brooks Holtom says outright that organizations spend little time analyzing workflows and understanding where exactly a human adds value. What bothers me in this story isn't AI, but the speed with which decisions affecting tens of thousands of people were made on the basis of presentations. Our assessment: the "AI boomerang" is first and foremost a diagnosis of management, not of the technology.

What this changes in practice. First, the price. Gartner believes rehiring will come at a premium: layoffs destroy talent pipelines and institutional knowledge. Second, negotiating leverage. It's now harder to convince someone invited back to an eliminated position to return at their old salary — and the advice in the AP piece sounds exactly like that: ask for more, training, clear prospects, and, per Holtom, a one-year renewable contract with a buyout clause if the job is terminated early. Third, "invisible work": Betsy Summers of Forrester reminds us about reading emotions, unplanned meetings, trust, and team cohesion — all the things agents can't deliver.

Who wins. Definitely not the shareholders who were sold the "right-sized" layoffs: Beaumont of SHL says outright that almost no one announces the reversal publicly. The winners are those who came back with a stronger position, and the companies that built AI from the start as an augmentation of people rather than a replacement. The losers are those who cut a role entirely and are now paying twice — for the implementation and for the return. We believe the labor market here isn't "canceling AI" — it's learning to price in the cost of a mistake.

⚠️ What We Don't Know

The biggest gap is scale. Neither AP nor Robert Half has an exact count of "boomerangs": 32% is a self-report from hiring managers, not accounting data. The same firm's previous survey put the figure at 29%, and a three-point difference is easily explained by sampling rather than a trend. Companies rarely confirm rehires publicly — which means part of this story we know only from the words of recruiters and analysts. Gartner's forecast of 30% by 2029 is a forecast, not a measurement. One more thing: in the list of examples, Microsoft, Barclays, and Kroger are mentioned as companies that cited AI in their layoffs, but the piece contains no specific rehiring figures for them.

Prediction: by the end of 2027, the share of rehires in Robert Half surveys will rise above 32%: the phenomenon will expand rather than fade, because the savings from layoffs run into the same hard cases that AI can't handle.

Disclaimer: The scale of the "AI boomerang" is not officially confirmed: neither AP nor Robert Half has exact rehiring figures, 32% is a self-report from 2,000 hiring managers, companies rarely admit rehires publicly, and the 30% by 2029 is a Gartner forecast, not a measurement.

Source: Associated Press