AI was sold to workplaces as a productivity booster. But a new survey shows the technology is creating a fresh problem for the people who manage employees. More than half of managers say they find it harder to evaluate their employees' work after adopting AI, according to the survey.
That finding cuts against the simple story that AI makes everything faster and easier. The people who have to judge the quality of work are now the ones struggling most.
Why Managers Struggle to Evaluate AI-Assisted Work
The survey points to a basic problem: when employees use AI to do their work, managers lose a clear view of what the employee actually contributed. Judging output used to mean looking at the finished product and the effort behind it. With AI in the mix, that line is much harder to draw.
The headline finding is blunt. More than 50% of managers report that evaluating employees' work has become harder since their teams started using AI. That is not a small group of holdouts. It is a majority of managers saying the tool has added friction to one of their core jobs.
The Productivity Promise Meets the Management Reality
AI promises productivity. For managers, the survey suggests, it is mostly a headache for now. The gap between the promise and the reality is the story here.
- More than half of managers say evaluating employee work is harder after AI adoption, per the survey.
- The difficulty is tied to judging work that employees produce with AI assistance.
- The survey frames AI as a productivity promise that has not yet delivered for managers.
To put it plainly, the survey does not say AI has failed. It says the management side of AI adoption is not working smoothly yet. The people who review, assess, and sign off on work are the ones feeling the strain.
What This Means for Teams Using AI at Work
For employees, the survey is a signal that using AI at work may not be judged the same way as doing the work yourself. For managers, it means the old ways of reviewing output may not fit a workplace where AI is part of the process.
The survey does not offer a fix. It simply records where managers stand right now: a majority finding their evaluation job harder, not easier, after AI arrived.
Our Take: The AI Productivity Story Has a Management Blind Spot
In our view, this survey exposes a blind spot in how AI is sold to companies. The pitch is always about speed and output. Almost nobody talks about the manager who has to look at that output and decide whether it is good.
That is the real headache. AI can generate work quickly, but it does not hand managers a clear way to measure who did what and how well. Until that gap is closed, the productivity promise will keep running into the same wall: the people in charge of judging work are the least equipped to judge it.
Readers who manage teams should take this seriously. If more than half of managers are already finding evaluation harder, the problem is not rare or temporary. It is the new normal of AI-assisted work, and it needs a real answer, not another productivity slogan.