Study links fear of AI career loss to more self-reported workplace misconduct

Workers worried that AI could make their occupation obsolete were more likely to justify breaking workplace rules and report counterproductive behavior, a Washington State University-led study found. The university described the findings on October 14, highlighting a workplace risk tied to fears about a career’s future—not simply the loss of one job.

The research, published in Occupational Health Science, examined workers across 18 industries. Researchers combined surveys about attitudes and behavior with the AI Industry Exposure index, an existing database measuring the potential for AI to displace tasks. They examined workers’ personal concerns alongside this industry-level measure.

The findings describe associations, not proof that AI adoption caused misconduct. The behavior measures were self-reported: they capture what employees said they did, rather than independently observed incidents of workplace wrongdoing.

When another employer is not the answer

Lead author Tahira M. Probst, a psychology professor at WSU Vancouver, distinguishes occupation insecurity from concern about a particular employer. A bartender whose bar closes can look for work at another bar, she explained. Her team examined a different fear: that the profession itself will disappear, leaving no equivalent role elsewhere.

Workers more worried about their occupation’s future reported more moral disengagement—the process of making behavior they would normally consider wrong seem acceptable. Those employees also reported more counterproductive conduct, including surfing the web and arriving late for shifts.

The authors interpret this pattern as a possible change in how threatened workers view workplace obligations. Employees may rationalize cutting corners or undermining others as self-protection, or as compensation for something they consider unfair.

Industry exposure sharpened the relationship

Finance and insurance had some of the highest AI exposure levels in the study, while agriculture and forestry had some of the lowest. These categories describe potential disruption to tasks, not measured AI adoption.

The relationships involving occupation insecurity, moral disengagement and counterproductive behavior were stronger in more exposed industries, including finance and insurance. They were less pronounced in fields facing fewer expected AI changes, such as forestry and construction.

The costs can reach coworkers

Probst emphasized that counterproductive behavior can damage more than an employer’s bottom line. Wasted resources and actions that interfere with other employees’ tasks can weaken workplace relationships and the organization’s ability to function.

The university recommends that employers pay attention to how they discuss and implement AI, keeping employees engaged and supported. That is a recommendation for managing the transition, not a tested intervention shown to reduce misconduct.