AI is giving employees time back, but many organizations are creating incentives for them to pretend it hasn’t.
Leaders want employees to transform the way they work. At the same time, many still reward people who look continuously occupied with today’s tasks. Those two expectations work against each other.
AI transformation requires available capacity. An organization operating with no room to spare cannot meaningfully adapt. It can only keep executing the work it already knows how to perform.
Businesses understand this principle elsewhere. Companies maintain financial reserves because unexpected events happen. Technology teams design extra capacity into important systems because running permanently at maximum load makes them vulnerable.
Yet when it comes to people, many organizations continue to behave as though every available minute should be occupied. That mindset leaves little space for learning, experimentation, reflection or transition. It also teaches employees to hide the time they save, because making that capacity visible often means it will immediately be filled.
The outcome is an unusual organizational performance: companies invest in AI to increase efficiency, employees use it to finish work faster, and then both sides maintain the fiction that the work still takes exactly as long as it did before.
What Happens When Greater Efficiency Only Produces More Work?
AI is reducing the time employees need to complete existing tasks. But when organizations respond by filling every newly available minute with another assignment, workers quickly discover there is little personal benefit to becoming more efficient. Saving an hour does not give them an hour to use differently. It simply creates another hour that can be filled with work.
The traditional employment model was built around the idea that employers were effectively purchasing employees’ time. That legacy helps explain why many organizations assume that whenever technology frees part of that time, it should immediately be reclaimed for additional output.
The logical response for employees is to conceal the hours they have saved and continue to look occupied. That appears to be exactly what many workers are doing.
A recent Software Finder survey found that 66% of U.S. respondents said they remain online or make themselves appear active after finishing their work. Those who did so reported spending an average of nearly five hours each week maintaining the appearance of productivity.
Perhaps even more concerning for managers, 64% said they had intentionally slowed down their work to avoid finishing too early because completing tasks quickly led to higher expectations.
Employees, in other words, are hiding their efficiency. Instead of using the time to learn a new capability, improve a process or prepare for what the business will need next, they move their mouse. They keep documents open or delay sending a response so they can still appear occupied later.
The Software Finder survey also found that among employees at companies using productivity-monitoring tools, 63% said monitoring made them more likely to fake activity.
The behavior extends upward through the organization. The same survey found that 73% of managers admitted they had also faked productivity for their own bosses. Whatever incentives are producing this response among employees appear to be shaping management behavior as well.
When respondents were asked what they would do if there were no consequences for finishing early, 71% said they would simply log off. Give people permission to stop when the work is complete, and most will stop rather than create extra activity to fill the remaining hours. Across organizational levels, employees are responding logically to the incentives they are given.
The pattern is not limited to the U.S. An Indeed survey of hybrid office employees in Germany found that two-thirds had deliberately taken steps to look more productive or engaged. More than one in four had artificially maintained an active online status, while 56% believed their employer placed more value on presence than measurable outcomes.
This is what happens when the measure itself becomes the goal. Employees learn to generate whatever signal the organization has decided matters.
If presence is tracked, people stay visible. If messages are counted, they send messages. If completed tasks are measured, they break work into more tasks. If AI consumption becomes a target, they consume more tokens.
The organization gets more visible activity. That does not necessarily mean it gets more value.
Why AI Productivity Metrics Can Misinterpret Employee Value
The same proxy measures that encourage employees to manufacture activity can also influence much more serious decisions about performance and employment. That risk came into focus in a recent lawsuit brought by 26 Meta employees in connection with company layoffs.
The employees allege that Meta relied on internal AI systems, activity data, AI-token-usage dashboards and algorithmically assisted performance information when identifying employees for layoffs. They argue that workers on protected medical, parental or family leave were unable to generate the same signals and were therefore placed at a disadvantage.
Meta disputes the allegations, stating that workforce decisions were made by people using documented and neutral criteria and that AI did not determine who was terminated. The case has not established that the systems in question actually determined the layoff list.
Regardless of the eventual legal outcome, however, it raises a larger workplace question: what kinds of information are managers beginning to associate with valuable work?
When a manager receives information that has already been organized around visible activity, recorded output or AI usage, the measurement system has shaped the definition of contribution before the manager begins evaluating anyone. Managers can override individual results, but they are still operating within an overall picture of employee value created by the available data.
Keystrokes tell a company that someone typed. AI-token consumption proves that someone used AI. Neither reveals whether AI improved the result, reduced the time required or simply created more unnecessary work. A high volume of messages does not prove influence. A packed calendar does not demonstrate meaningful contribution. A large digital footprint may suggest productivity, but it can just as easily reflect inefficient processes, excessive meetings or work that never needed to be done.
One employee can generate enormous amounts of visible output while solving the wrong problem. Another can prevent a costly mistake during a short conversation that creates almost no measurable digital trace.
When organizations measure work poorly, employees eventually change their behavior to produce whatever the company has decided to count.
How Organizations Should Use The Time AI Creates
AI was supposed to create capacity for more valuable work, not simply allow organizations to fit more of today’s work into fewer minutes. Achieving that requires organizations to slow down in some areas in order to move faster later.
Employees need time to understand how their roles are changing. They need space to test new tools, learn where those tools fall short and redesign the processes into which the technology is being introduced.
They also need time to build the capabilities required for work the company may not yet perform. The World Economic Forum’s Future of Jobs Report found that employers expect nearly 40% of the skills required in the workplace to change by 2030. Skills gaps are already the most frequently cited barrier to business transformation among the employers surveyed. The report estimates that 59 out of every 100 workers will require training by 2030.
Where is that learning supposed to happen if every minute saved through AI is immediately assigned to additional output?
Some of the time created by AI also needs to remain available for work that does not generate an immediate or easily measured result: improving quality, better serving customers and addressing problems that have been neglected.
Measure What The Business Will Need Next
Organizations can generate just as much busywork by measuring the wrong outputs as they can by monitoring the wrong activities. What ultimately matters are outcomes: whether the work moved the organization closer to its goals. But those outcomes also depend on future capability, meaning the organization’s ability, and its employees’ ability, to perform the work that will be needed next.
Future capability is easy to overlook precisely because its value appears later. It may not show up in the next financial cycle. Learning a new platform, redesigning a workflow or building better judgment can temporarily reduce visible productivity. Without that investment, however, a company may become extremely efficient at performing work that is becoming less relevant.
Before leaders increase productivity expectations because AI has made work faster, they need to decide how they want to use the capacity AI creates.
How much of that time should go toward improving customer outcomes? How much should increase quality? How much should be used to redesign work? And how much should be invested in learning, experimentation and the capabilities the organization will require in the future?
If leaders do not answer those questions deliberately, the default response will almost always be to add more work.
And employees will keep getting better at looking busy.
