Asger Risom
AI tools are increasing cognitive load for many workers, with research suggesting they may be making jobs more mentally demanding rather than less.
AI was supposed to make work easier. Instead, some researchers say it may be doing the opposite.
According to Fortune, a growing body of research suggests that using AI tools is increasing cognitive load, leaving workers more fatigued, more error-prone, and struggling to keep up with rising expectations.
Productivity gains are being offset by mental strain
The issue isn’t just the technology itself—it’s how work is changing around it.
Studies show that when routine tasks are automated, workers are left with more cognitively demanding responsibilities.
At the same time, AI enables higher output, which often leads to more work being assigned rather than less.
An eight-month study of 200 employees found that AI use intensified workloads instead of reducing them. Separate research has described a “brain fry” effect, where combining AI usage with existing responsibilities leads to worse performance, not better.
The brain isn’t built for constant AI interaction
Part of the problem comes down to human cognitive limits.
Working memory can only handle a small amount of information at once—often as few as three to five items.
Add frequent task switching, such as moving between AI prompts and real-world tasks, and the strain increases.
Research suggests it can take more than 20 minutes to fully regain focus after switching between tasks. In AI-heavy workflows, that recovery window is constantly disrupted.
Over time, that leads to fatigue, missed details, and declining performance—even among high-performing employees.
More tools, more work—not less
AI also creates a paradox: the more capable the tools become, the more work people are expected to handle.
Instead of replacing tasks, AI often adds a layer of oversight—reviewing outputs, refining prompts, and managing results. For many workers, that means longer hours and a higher mental load.
The result is not less work, but a different kind of work—one that is often more demanding.
What leaders are being told to change
Researchers argue that companies need to rethink how AI is implemented.
One recommendation is to create dedicated “quiet time” without meetings or AI use, allowing employees to focus deeply or step away mentally.
Another is to shift performance measurement away from hours worked and toward outcomes.
There is also a growing emphasis on training employees to use AI effectively—not just as a shortcut, but as a tool that enhances thinking rather than replacing it.
A tool that still needs guardrails
AI is not inherently harmful, but its current use may be outpacing how people actually work.
The core issue is not the technology itself, but the mismatch between human cognitive limits and how AI is being deployed in real-world environments.
Without adjustments, the tools designed to boost productivity risk doing the opposite.
Sources: Fortune, research by the NeuroLeadership Institute, study by Aruna Ranganathan and Xingqi Maggie Ye, BCG research
