gensai.aiAI rebounding describes a company backtracking after replacing employees with AI, then having to call them back because the tool does not do the expected work. The assessment firm SHL gave the phenomenon its name on 29 October 2026: in its survey of 2,036 workers carried out in June and July 2026, 19% of US respondents and 13% of UK respondents said they had been replaced by AI and later asked back by the same employer. Gartner predicts that by 2029, 30% of employees terminated and replaced by AI will be rehired, often at a higher cost.
In practice at Gensai
Gensai designs its projects to support teams rather than replace them: AI handles simple, repetitive requests and passes complex cases on to a person. Projects usually start with a POC that measures the real result before any change to the organisation.
How to avoid AI rebounding?
By measuring what AI does on real work before touching the organisation: a pilot of a few weeks, tracked indicators (requests resolved, errors, satisfaction) and a period where the tool and the team work side by side. Gartner advises starting from the work to be done rather than from a headcount target: in its view, only 19% of organisations have workforce planning that accounts for AI.
Common mistake
Deciding how many roles to cut before knowing what AI really does. Klarna started hiring human customer service agents again in 2025 after betting on its AI assistant, and Ford recruited 350 experienced engineers in June 2026 when its AI quality-control tools fell short. What the tool lacks is often the know-how of the people who left: difficult cases, exceptions, customer relationships.
Same theme
Definition written with AI assistance, checked by Sébastien Lecocq, co-founder of Gensai.
Going beyond the definition?
From concept to project: Gensai builds custom, sovereign, GDPR-compliant AI solutions.
Related services: AI agents, Custom AI solutions