Nearly a third of US hiring managers say their organisations have brought back jobs they previously eliminated because of AI or automation, according to a survey of 2,000 managers by the staffing firm Robert Half. The 32 percent figure describes employers restoring the exact roles they had cut, a reversal now common enough that recruiters have a name for it: the "AI boomerang." The firm said the rehiring spans customer service, engineering, finance, marketing, law and health.
Managers who rehired gave consistent reasons. Forty percent cited organisational knowledge and context beyond what AI could handle, 38 percent said the tools needed more supervision or quality checking than expected, and 35 percent said productivity gains fell short of what was promised. "When you need critical thinking, collaboration, creativity, institutional knowledge, those are things that AI just can't replicate," said Bill Driscoll, a senior district president at Robert Half. Analysts describe the common failure as missing the "invisible work" employees do beyond their formal job descriptions.
Two named cases illustrate the pattern. Klarna, the Swedish buy-now-pay-later company, began rehiring human customer service agents last year after chief executive Sebastian Siemiatkowski acknowledged its push to replace them with a chatbot had gone too far; it now routes routine queries to automation and harder cases to people. In June, Ford said it had hired or promoted about 350 veteran engineers after automated quality checks underperformed, crediting the combination of AI and human expertise for later quality gains.
The reversals do not mean AI-driven cuts have stopped. Microsoft, Barclays and the supermarket chain Kroger are among companies that announced AI-related workforce reductions this year, and the cuts are affecting tens of thousands of people. What the Robert Half data suggests is that some of those decisions were made before employers understood which tasks automation could reliably take over. Lucy Beaumont of the testing firm SHL noted that companies rarely advertise the U-turns, sometimes admitting "quietly, with their tail between their legs" that a round of cuts was premature.
The finding fits a muddier jobs picture than either boosters or doom forecasts predicted. US professional and business services added 70,000 payrolls in October, well above the sector's recent monthly average, which one economist called "a good reminder that AI isn't wiping out white-collar jobs." AI Daily reported last week that US employers had cited AI in roughly 101,000 job cuts in the first half of 2026 even as the broad white-collar collapse failed to materialise. The boomerang adds a further wrinkle: the question is not only whether AI eliminates jobs, but whether the eliminations hold. Robert Half did not say how long rehired roles stayed filled, or how many cut positions were never restored, so the survey measures the frequency of reversals rather than the net effect on employment.