Study Warns: Managing AI Tools Exhausts Workers, Leading to High Turnover

Among those utilizing AI at work, individuals experiencing AI brain fry were 11 percent more likely to commit minor errors and 39 percent more prone to major mistakes compared to their peers who did not exhibit these symptoms. Furthermore, 34 percent of those affected expressed an intention to leave their jobs, in contrast to only 25 percent of those who did not report such fatigue.

The prevalence of AI brain fry varies significantly across different job roles. For example, just 6 percent of legal professionals reported experiencing it, while 26 percent of marketing professionals faced similar issues. Other roles in human resources, operations, engineering, finance, and IT also reported high instances of this cognitive exhaustion.

This study’s findings emerge during a period of aggressive AI implementation in businesses, often driven by top management.

Nvidia CEO Jensen Huang has previously encouraged employees to “use AI for every task.” Many organizations are further incentivizing AI use by tracking and rewarding token consumption as an indicator of employee performance. Recently, Meta reportedly incorporated AI-generated code lines as a metric for engineers’ productivity. This larger push for AI integration has coincided with growing employee concerns over potential job losses.

“Our findings serve as both a guide and a caution. When applied thoughtfully, this data can inform the development of AI-supported workflows that mitigate burnout. They also highlight specific managerial, team, and organizational approaches for avoiding mental fatigue even as AI usage escalates,” the researchers remarked in an article published in the Harvard Business Review on March 5.

Key Findings of the Study

Oversight of AI Tools: The study reported that the oversight required to use AI tools effectively demands an additional 14 percent of mental effort from workers. This heightened involvement results in 12 percent more mental fatigue and 19 percent greater information overload among respondents.

Number of AI Tools: It was noted that using an excessive number of AI tools can hinder productivity. Participants who employed more than three AI tools indicated a decline in their productivity levels.

Mental Fatigue vs. Burnout: Interestingly, while engaging with AI tools increased mental fatigue, the automation of repetitive tasks through AI led to a decrease in overall burnout levels. The study clarifies that burnout encompasses broader physical and emotional stress, distinct from AI brain fry, which is characterized by acute cognitive strain due to heavy AI use.

Business Costs of AI Brain Fry

The study attempts to quantify the repercussions of AI brain fry, stating that the cognitive strain from intensive AI usage incurs significant business costs, notably the impact on decision-making quality.

Participants experiencing AI brain fry reported 33 percent more decision fatigue than those not affected by it. Referencing a 2018 study, the researchers noted that suboptimal decision-making at a company with $5 billion in revenue can cost approximately $150 million each year. Consequently, a 33 percent increase in decision fatigue could result in millions of dollars in added expenses annually, as stated by the researchers.