Is AI Making an Existing Workplace Crisis Worse? EAPs Can Help
Key Takeaways
- Artificial intelligence arrived at a time when attention spans had been in decline for over a decade.
- Getting real value from AI depends on employees reviewing its output carefully, which requires conditions many workplaces don't currently provide.
- Two well-established research areas, attention residue from task-switching and automation complacency from reviewing automated output, offer plausible explanations for how AI-assisted work could strain attention further.
- Organizations that protect employee focus as infrastructure, not an individual responsibility, are better positioned to capture AI's value rather than just adopt it quickly.
Artificial intelligence burst onto the scene just a few years ago, transforming many workplaces and even entire industries, and creating valid concerns for both employers and employees.
- Among U.S. employers using or planning to use AI, 89% anticipate workforce implications.
- Likewise, 44% of workers worry that AI and robotics would make their skills unnecessary.
Yet, there has been another, perhaps even more important change that predates AI: declining attention spans.
Research beginning in 2003 found that people spent an average of about two and a half minutes on a computer screen before switching. By 2012, the same research approach found an average of about 75 seconds. In more recent studies, the average was about 47 seconds.
That data points to something significant. Sustained focus on any one thing has been getting harder to come by.
The decline in attention spans occurred during the rise of smartphones, open offices, always-on messaging, and a workplace culture that treats availability as a virtue. Today, generative AI is arriving into that environment, creating additional reasons for concern.
What makes this matter for HR leaders and executives is that getting real value from AI depends on employees reviewing its output carefully, catching what it gets wrong rather than accepting the first answer. A workplace already struggling with sustained focus is poorly positioned to do that reviewing well, regardless of the tool.
Attention is the Foundation of Mental Fitness
Mental fitness describes an employee's capacity to focus, manage stress, and think clearly under pressure. It draws on several things researchers have studied separately for years, including cognitive load and the ability to recover between periods of demanding work.
Attention sits underneath all of it.
The decline in attention exemplified by screen switching didn't emerge from a single cause. What's clearer, though, is the pattern itself. The workday increasingly rewards constant availability over sustained focus.
That pattern carries a real cost. Gloria Mark, the researcher at UC Irvine who tracked the attention shift, has also studied what happens after an interruption. Her field research found that people often take upward of 25 minutes to return to an interrupted task, frequently completing two or more other tasks in the meantime.
This does not mean employees are unable to think during the entire interval. It means interruptions can reshape the workday, as people address other requests and responsibilities before returning to the original priority. The practical consequence is that resuming focused work often takes much longer than employees and managers expect.
Technology is Advancing Faster Than Human Capacity
While other technologies have had significant impacts on the workplace, artificial intelligence is bringing something else entirely, in terms of both capabilities and speed.
Stanford’s 2026 AI Index Report estimates that generative AI reached 53% population-level adoption within three years of its mass-market introduction, a faster pace than the personal computer or internet reached over comparable periods.
Given the relative newness of AI, research into its effects on attention is still in its infancy, so claims that it is actively shrinking attention spans go further than the evidence supports. However, what does exist is well-established research on two related mechanisms, both of which offer plausible extensions to how people use it.
What Happens When AI Interrupts Unfinished Work
In 2009, researcher Sophie Leroy documented something she called attention residue.When someone switches away from an unfinished task, part of their focus stays behind, and it measurably hurts their performance on whatever comes next. The effect is strongest when the original task was left incomplete.
Pausing a task to prompt AI, wait for a response, and decide whether to use it may create a similar switch, particularly when that detour pulls the employee into a different task or leaves the original problem unresolved. Using AI as a bounded step within the same task may not trigger the same effect.
What Happens When AI Turns People Into Reviewers
A separate, decades-old body of research in aviation and industrial automation found that people monitoring an automated system get measurably worse at catching its errors over time, particularly when that monitoring competes with other demands on their attention.
Researchers call this automation complacency. It shows up in both novices and experts. Experience alone doesn't resolve it, though training, clear accountability, and better verification procedures can help.
AI use can put people in a similar reviewing role, as evaluators of output that someone (or something) else produced.
It is too early to assume that findings from cockpits and control rooms apply directly to AI-assisted knowledge work. Still, the underlying lesson is worth taking seriously: sustained monitoring and error detection are demanding skills, particularly when employees face competing demands on their attention.
The Next Competitive Advantage is Mental Fitness
Some companies will get far more value from AI than others, and the difference may not simply come down to who has the most talented workforce or the most advanced tools.
Automation complacency research offers a clue why. Getting real value from AI isn't purely a matter of individual skill or effort. The environment around an employee, whether it protects focused time for review or fills every hour with competing demands, shapes outcomes as much as anything else.
Consider two employees using the same AI tool. One works in an environment with back-to-back meetings, constant messaging pings, and no real expectation that reviewing AI output takes real time. The other has protected blocks without interruptions, and a manager who treats a careful second look at AI-generated work as part of the job, not a delay to route around. Both employees may have identical AI training. Only one of them is actually positioned to catch what the tool gets wrong.
The real competitive question is which company builds an organizational culture where people can still pay real attention to what AI produces, separate from which company adopts AI fastest.
Organizations Must Build Mental Fitness, NOt Just Expect It
Building that kind of environment takes more than good intentions. The World Health Organization's workplace mental health guidelines name the structural conditions that undermine focus and well-being: unclear roles, low control over how work gets done, and excessive workload.
None of these are new, and none of them require AI to matter. AI raises the cost of leaving them unaddressed.
4 Ways HR Can Help
-
Protect blocks of uninterrupted time on employees' calendars. Reviewing AI output carefully takes real, focused time, and back-to-back meetings with constant notifications don't leave room for it.
-
Set explicit expectations that reviewing AI-generated work is real work, not friction to route around. When speed is the only thing rewarded, careful review is the first thing employees cut.
- Structure AI use around clear task boundaries. Encourage finishing a task, or reaching a clear stopping point, before switching to prompt AI mid-task, which may limit the kind of unfinished-task switching linked to attention residue.
- Train managers to model attentive review, not just AI adoption. Experience alone doesn't build strong review habits, but clear expectations and visible modeling from managers can. Treat this as a trainable skill, not something employees are left to develop through individual discipline.
Treating attention as a resource worth protecting puts it on the same footing as any other asset central to how well an organization's people actually perform.
Rethinking the Role of Employee Assistance Programs
EAPs are often the first resource organizations reach for when addressing employee mental fitness, and there's good reason for that. They're relatively fast to deploy, and they give employees direct access to real support, such as confidential counseling, referrals, and in many cases, manager consultation when a situation calls for it.
That support has genuine value for the individual employee using it.
However, a counseling session doesn't reduce the number of interruptions in someone's day. A referral doesn't give an employee more control over whether they're expected to review AI output carefully or just move fast. Manager consultation doesn't redesign a workload that was unsustainable before AI ever entered the picture.
Individual support and structural change solve different problems.
EAPs sometimes get positioned, implicitly or explicitly, as the organizational response to workplace mental health concerns. They work better as one layer of a broader strategy, alongside changes to workload, interruption culture, and how AI use gets structured. An employee who gets excellent individual support and then returns to the same fragmented, interruption-heavy environment is likely to end up back where they started.
The distinction is a useful one for HR leaders evaluating their own mental health strategy. Is the organization treating EAP access as the plan, or as one part of a plan that also includes protected focus time, clear expectations, and deliberate AI use?
The first approach supports individuals. The second addresses the source of the strain directly.
Investing in People is the Best AI Strategy
The decline in sustained attention isn't a problem AI created, and it won't disappear if AI adoption slows down. It has its own long history, and it will outlast whatever specific technology comes next.
Organizations that protect attention stand to gain something broader than better AI outcomes: a workforce that can focus, absorb change, and think clearly under pressure, regardless of what tool they're using.
AI adoption makes that investment more urgent, since reviewing AI output well depends on exactly the kind of sustained focus that's already in short supply. But the investment itself, protected time, clear expectations, managers who model careful attention rather than constant reactivity pays off well beyond AI. It's what lets a workforce handle demanding work of any kind.
When you partner with Ulliance, our Life Advisor Consultants are always just a phone call away to teach ways to enhance your work/life balance and increase your happiness. The Ulliance Life Advisor Employee Assistance Program can help employees and employers come closer to a state of total well-being.
Investing in the right EAP or Wellness Program to support your employees will help them and help you. Visit https://ulliance.com/ or call 866-648-8326.
The Ulliance Employee Assistance Program can address the
following issues:
• Stress about work or job performance
• Crisis in the workplace
• Conflict resolution at work or in one’s personal life
• Marital or relationship problems
• Child or elder care concerns
• Financial worries
• Mental health problems
• Alcohol/substance abuse
• Grief
• Interpersonal conflicts
• AND MORE!
References:
The 2026 AI Index Report; Stanford Institute for Human-Centered Artificial Intelligence
https://hai.stanford.edu/ai-index/2026-ai-index-report
Can't Pay Attention? You're Not Alone; University of California; Cara Capuano
https://www.universityofcalifornia.edu/news/cant-pay-attention-youre-not-alone
Complacency and Bias in Human Use of Automation: An Attentional Integration; Human Factors; Raja Parasuraman and Dietrich H. Manzey
https://journals.sagepub.com/doi/10.1177/0018720810376055
Mental Health at Work; World Health Organization
https://www.who.int/news-room/fact-sheets/detail/mental-health-at-work
Why Is It So Hard to Do My Work? The Challenge of Attention Residue When Switching Between Work Tasks; Organizational Behavior and Human Decision Processes; Sophie Leroy
https://ideas.repec.org/a/eee/jobhdp/v109y2009i2p168-181.html