Until recently, it seemed like a simple plan. Companies laid off some of their employees, handed their work over to artificial intelligence, and started calculating the savings. But many of them are now finding that it does not work so simply in practice and are contacting former employees with offers to return. They are often willing to come back, but no longer under the same conditions as before.
A new study by employee assessment company SHL shows how widespread this phenomenon is. According to the survey conducted in June and July, more than 19% of U.S. respondents said that they had personally been replaced by artificial intelligence and were later asked by their employer to return. Across the U.S. and the UK combined, the figure was 16%.
Experts use the term “AI rebounding” for this phenomenon. SHL uses it to describe a situation in which an employee is replaced by AI and then asked by the same employer to return.
One in five received an offer to return
The Talent Mobility at Work in 2026 report is based on a survey of more than two thousand working adults in the United States and the United Kingdom. The survey was conducted for SHL by Propeller Insights, and the company published the results in late October.
The data shows that this is not a marginal issue. Slightly more than half of people had encountered employees returning after the introduction of AI, whether personally, through someone they knew, or by hearing about such cases. The differences between countries are noticeable. While nearly one in five workers in the U.S. has direct experience, the figure in the UK is 13%.
Young people are affected the most. In the 18-to-24 age group, 24% had personally experienced such a return, while 36% knew someone with the same experience. This makes sense because companies often try to automate entry-level roles, such as customer support or simpler office work, first.
SHL’s findings fit into a broader picture. In recent months, artificial intelligence has become one of the most common reasons companies give for layoffs. According to consulting firm Challenger, Gray & Christmas, employers cited AI as the primary reason for nearly 40% of the layoffs announced in May 2026. But the wave of layoffs was followed by a wave of rehiring. A March report by workforce planning company Orgvue found that 32% of organizations that had laid off employees because of AI subsequently had to hire again.
Why AI failed to do what people could
According to SHL, the problem is that companies often act too hastily. “AI rebounding happens when a company automates a role before it has truly assessed the human capabilities embedded in it,” explains Lucy Beaumont, SHL’s vice president of product and design. According to her, artificial intelligence then fails to deliver what was expected of it, the gap suddenly becomes apparent, and the company ultimately rehires the same experts it let go, but at a much higher cost.
Beaumont points out that a deeper problem lies behind this. According to her, most companies do not have a reliable and objective overview of what their people are actually capable of. Management therefore sees only a job description, not the experience and subtle skills that a person has accumulated over years on the job. According to Beaumont, artificial intelligence cannot fully replace employees who understand customers, know how to make the right decisions, show empathy, and maintain relationships with people.
Specific cases show that this is not merely theoretical. Major companies such as Amazon, Ford, and Klarna have already brought employees back after introducing AI because their artificial intelligence plans did not work out as expected. Automaker Ford, for example, rehired engineers it had believed could be replaced by artificial intelligence.
The best-known example is Swedish fintech company Klarna. Its CEO, Sebastian Siemiatkowski, boasted to shareholders that the company’s AI assistant was doing the work of 700 employees. In the end, however, he had to admit that this had led to poorer service quality and lower customer satisfaction, and that human contact was still necessary. Klarna therefore began recruiting customer support staff again.
Regret over layoffs is not uncommon. According to an earlier Orgvue report, 39% of companies laid off employees because of automation, and 55% of them now regret it. Figures from staffing agency Robert Half paint a similar picture. Its data shows that 32% of U.S. hiring managers eliminated a position primarily because of AI and later hired again for the same or a similar role.
Analysts expect the number of returns to increase. Consulting firm Gartner predicts that by 2029, 30% of people laid off or replaced by artificial intelligence will return to work, often at a higher cost.
People set their terms when returning
People sent home by a company because of artificial intelligence are not so bitter that they immediately reject every offer to return. At the same time, however, they are well aware that they now hold stronger cards. According to SHL, most replaced workers are willing to return, but not under the same conditions.
Money comes first. A total of 51% of people would want a significantly higher salary before returning, 42% would demand specific guarantees that they would not lose their jobs again, and 40% would want complete transparency about how artificial intelligence would be used in their work. The last demand is understandable. Anyone who has once lost a job because of an algorithm wants to know in advance whether it could happen again.
Companies should expect returning employees to be more demanding. Beaumont points out that people who have been laid off because of AI will expect more from employers than they did before. The savings that management promised shareholders can therefore easily turn into the opposite. The company first pays to implement the technology and provide severance, then pays for the loss of quality, and finally pays even more to bring back the same people.
Only a small proportion of those laid off would not return at all. One worker in thirteen, or roughly 8%, would refuse to return under any circumstances. Among Americans, the figure is just 6.6%. Most people are therefore not closing the door; they are simply asking for a better price to open it.
Sources: fortune.com, cnbc.com and timesofindia.indiatimes.com
