UK workers say some employers replaced staff with artificial intelligence, then asked them to return, according to a survey commissioned by talent-assessment company SHL. The findings describe respondents' accounts, not a verified count of UK redundancies reversed after AI failed to meet expectations.
13% of UK Respondents Reported Being Asked Back
Propeller Insights conducted the online survey in June and July 2026, questioning 2,036 employed adults in the UK and US. The sample included 1,007 UK respondents and 1,029 US respondents.
Thirteen per cent of UK respondents said they had personally experienced what SHL calls 'AI rebounding': being replaced by AI and later asked back by the same employer. Across both countries, 16% reported the experience.
The survey records respondents' accounts, not independently checked employment histories. It does not establish how many UK employers reversed redundancies, whether AI alone prompted the original decision, how many workers accepted a return offer or what terms they received.
A reported sequence does not prove an AI system could not perform a role.
That limit matters to the headline claim. Lucy Beaumont, SHL's global senior vice-president of product and design, said in the company's announcement that AI 'doesn't deliver' and employers rehire former staff 'at a far greater cost'. These are Beaumont's characterisation and SHL's interpretation; the survey provides no financial comparison of employment arrangements.
Workers Say They Would Seek Better Terms
The survey announcement says 51% of workers would want a substantial pay increase before returning after AI-related displacement. It reports that 42% would want specific guarantees about future job security and 40% would seek transparency about AI's future use in their roles.
The release does not clearly state whether these percentages apply only to respondents who personally experienced AI rebounding or to a broader group asked about return terms.
The figures describe stated expectations, not confirmed negotiations or pay rises. For a UK worker offered a return, they identify issues of compensation, job security and how AI would affect the role. The survey does not show whether employers met those expectations or whether anyone returned at higher cost.
Broader Research Measures Employment Trends
Other research tracks employment at a wider scale, but does not verify the SHL respondents' accounts. Stanford Digital Economy Lab researchers analysed payroll data covering millions of US workers through June 2026. They found no evidence of widespread, economy-wide displacement.
Employment among workers aged 22 to 25 in highly AI-exposed occupations was 19% below the level it would have reached if it had kept pace with similarly aged workers in less-exposed occupations.
The Stanford researchers said the gap operated primarily through reduced hiring of younger workers rather than increased separations. They described the results as 'early, descriptive indicators' rather than causal estimates, saying the data alone could not establish how much of the divergence was caused by generative AI.
The International Labour Organization's review says large-scale job displacement remains limited in the research examined. It also identifies risks involving younger workers and job quality. Neither source establishes the effect on UK workers specifically or measures how often employers ask former staff to return.
SHL's survey offers a snapshot of reported experiences and workers' stated expectations. Establishing how often UK employers reverse AI-related job cuts, and whether those reversals cost more, would require separate evidence from employers or employment records.
