AI news ·
Nearly one in five U.S. workers replaced by AI get asked to return, survey finds
19.1% of U.S. workers replaced by AI were later asked back by the same employer, a cycle SHL calls "AI rebounding." Among those rehired, 51% demanded a substantial pay increase to return.
Nearly one in five American workers have been replaced by AI only to be asked back by the same employer, according to survey data from talent assessment firm SHL. The findings point to an emerging trend the firm calls "AI rebounding" - a costly cycle where companies automate roles too quickly, discover the limits of the technology, and scramble to rehire the people they let go.
The survey, conducted in June and July, found that 19.1% of U.S. respondents said they personally experienced this boomerang effect. Across both the U.S. and the U.K., the figure was 16%. The data suggests some organizations are making workforce decisions based on an incomplete understanding of what AI can and cannot do.
"Organizations are making fast, sometimes myopic, decisions about their people based on an incomplete picture of what their employees can do with or without AI," said Lucy Beaumont, global senior vice president of product and design at SHL. "AI rebounding is what happens when a company automates a role before it has genuinely assessed the human capability inside it. The AI doesn't deliver, the gap becomes obvious, and the business ends up rehiring the same experts they just let go of, at a far greater cost."
Workers want more money and guarantees to return
The rehiring process is not cheap. Among workers who were asked to return, 51% said they would require a substantial pay increase before saying yes. Another 42% wanted written guarantees about future job security, and 40% demanded transparency about how AI would be used in their roles going forward.
The SHL data aligns with a separate Robert Half survey from July, which found that 32% of U.S. hiring managers who eliminated positions because of AI later reinstated the same or similar roles. The top reasons: institutional knowledge AI could not replace (40%), relationship management needs (39%), increased business demand (39%), and more human oversight required than expected (38%).
Klarna and Ford hit the limits of automation
Some well-known companies have already lived this pattern. Swedish buy-now, pay-later firm Klarna began recruiting human customer service agents again after leaning heavily on AI chatbots. CEO Sebastian Siemiatkowski acknowledged the company had gone too far in its push toward automation. A Klarna spokesperson said human agents were needed "for complex, sensitive cases" where the "human touch" brings the most value.
Ford took a similar step in June, hiring and promoting 350 experienced engineers after AI-powered quality control tools underperformed. The automaker said AI contributed to quality improvements but emphasized that the technology works best alongside seasoned engineering teams. A Ford representative told Fortune that "these engineers carry the hard-earned wisdom of decades of design."
Younger workers are more likely to be replaced - and rehired
AI-related layoffs are accelerating. Employers cited AI in 121,401 announced job cuts through October, according to outplacement firm Challenger, Gray & Christmas - more than double the 54,836 layoff plans announced for all of 2025. Younger workers face disproportionate risk. A Stanford Digital Economy Lab study found that employment among workers aged 22 to 25 in AI-exposed occupations was 19% below where it would have been if it grew at the same pace as their less-exposed peers.
Yet SHL's research shows a silver lining for that age group. Workers aged 18 to 24 were more likely to report experiencing AI rebounding: 24% said they had been replaced by AI and later asked to return, compared to 19.1% of U.S. respondents overall. The data hints that early-career professionals may still hold skills that prove harder to automate than first assumed.
Why this matters for HR leaders and executives
AI rebounding is a signal that workforce planning cannot be outsourced to a cost-cutting algorithm. When companies eliminate roles without first auditing the full scope of what their people contribute - relationship management, institutional memory, judgment in edge cases - they risk paying more to fix the gap later. For HR leaders, the takeaway is clear: skills assessment must precede automation decisions, not follow them. Professionals responsible for talent strategy can build that capability through structured learning, such as AI VP HR Courses or AI HR Leadership Courses, which focus on evaluating roles before deploying AI. The companies that get this right will treat AI as a tool that augments expertise, not a cheaper substitute for it.