Fear of AI Is Making Employees Slack Off — and Steal Lunches

Employees who fear artificial intelligence will take their jobs are more likely to slack off, arrive late, surf the web on company time — and even steal a coworker's lunch, according to a new Washington State University study published in the journal Occupational Health Science.

The research, led by Tahira M. Probst, a professor of psychology at WSU Vancouver, surveyed employees across 18 different industries. It examined how AI adoption — intelligent robotics, automated systems, and algorithmic management — changes worker attitudes and behavior on the job.

The pattern was consistent: the greater the fear of replacement, the worse the conduct.

What did the study actually find?

Researchers identified two distinct problems. First, workers with the strongest fears about AI stealing their jobs behaved poorly in direct ways — slacking off at the office, or in extreme cases taking a colleague's lunch. Second, workers who assumed they would be replaced were more likely to "morally disengage," a term for rationalizing misconduct such as showing up late for a shift or browsing the web while on the clock.

Probst explained the mechanism in a news release: "Moral disengagement is basically a way of justifying bad behavior. We know what's right and wrong. We know what organizational policies say we can and can't do. Moral disengagement is a way of reframing that behavior as something that is acceptable."

That reframing is the part employers should worry about most. It turns anxiety into self-permission.

The study's bottom line is blunt: when employees feel their job is at risk, they reinterpret workplace expectations and rationalize behaviors such as cutting corners or undermining coworkers — to protect themselves or to compensate for what they perceive as injustice.

Which industries are most exposed?

The researchers used the AI Industry Exposure index to rank sectors by potential disruption. The results split sharply:

• Highest exposure: finance and insurance
• Lowest exposure: agriculture, forestry, and construction

The bad behavior tracked the exposure numbers. Misconduct was more pronounced in industries facing the highest amount of AI disruption, the study found.

That matters for any manager rolling out automation in a finance-heavy environment. The same tool that cuts costs may quietly corrode conduct.

What does this mean for employers?

The findings carry a warning that cuts both ways. AI concerns harm not only the workers who feel threatened but also the business itself, the researchers concluded. Anxious employees who disengage morally cost the company productivity, discipline, and trust — before any robot arrives.

The study recommends that employers be more mindful of how they bring AI into the workplace so that employees are better supported. That is the promise in the report; what it does not spell out is which specific support measures actually work. The research establishes the link between AI anxiety and misconduct, but the remedy remains largely up to individual companies to design and test.

Prior research cited by the WSU team already showed that AI-driven technologies increase job insecurity, stress, and disengagement. The new study adds the behavioral layer: insecurity does not just feel bad, it changes what people do at their desks.

For firms racing to deploy AI, the message is that implementation is now a workforce-management problem as much as a technology decision. Companies that automate without addressing employee fear should expect the costs to show up not in the IT budget, but in lateness, corner-cutting, and missing lunches.