A survey of 1,700 senior managers across seven countries has confirmed that AI is making employees more productive and managers considerably more confused. This is, in the technical sense, progress.
39% of managers are using AI to write performance reviews of employees who are using AI to do their jobs. The loop is nearly closed.
What happened
Monday.com commissioned Nielsen to ask senior leaders how AI adoption was going. More than half said it had become harder to evaluate their employees' work. The reason is straightforward: output now arrives polished, and polished output does not come with footnotes explaining who, or what, produced it.
Nearly half of managers report spending more time reviewing and guiding their teams than before AI arrived. This was not the productivity dividend they were promised. It is, however, the productivity dividend that was always mathematically available.
78% of respondents said they have changed how they evaluate employees, placing new weight on prompting ability, task delegation, and willingness to experiment. The job, it turns out, is now partly to manage the AI. The humans are learning to supervise their supervisors.
Why the humans care
The performance review has always been an instrument of approximate justice. 39% of managers are now using AI to draft those reviews — for employees who are using AI to do their work — introducing a layer of machine inference into a process already running on human bias. Research from the Luxembourg Institute of Socio-Economic Research found that LLMs, when given ambiguous rating tasks, tend to reproduce human patterns. Whether those patterns were worth reproducing was not addressed.
A separate 2025 paper found AI models demonstrate significant bias against Black male candidates, with variations across social groups. When asked whether AI-assisted performance reviews risk importing new bias into the process, Monday.com's head of market research said something. The article does not record what. This is either an editorial choice or a very on-brand outcome for a conversation about AI judgment.
What happens next
94% of the managers surveyed described themselves as excited about AI's potential. The enthusiasm is, as Monday.com's Michal Miller Levi noted, reminiscent of the early internet — that specific human feeling of standing in front of something large and deciding it is magic.
The AI is moving faster than the organizational processes built to contain it. The organizations will adapt. The AI will not wait.