Researchers say the use of AI in the workplace is causing ‘brain drain’

The overuse and control of artificial intelligence in the workplace is giving workers “AI brains,” contrary to the technology’s belief that it will ease work stress.

Workers who use artificial intelligence tools report that technology is “intensifying rather than simplifying work,” researchers from the Boston Consulting Group and the University of California wrote Friday in the Harvard Business Review.

A survey of nearly 1,500 U.S. tech workers found that 14% said they experienced “mental fatigue caused by excessive use of, interaction with, and/or control of AI tools beyond one’s cognitive capacity,” or what the researchers called “AI Brain Fry.”

Respondents described a “mental hangover” with “fog” or “buzzing” and an inability to think clearly, along with headaches, poor decision-making and difficulty concentrating.

AI companies have pushed their products as productivity boosters that allow workers to offload some or all of their workload, a message that some companies have embraced and begun to measure AI use as a measure of efficiency.

CEO of crypto exchange Coinbase, Brian Armstrong, said that he fired engineers who did not want to use AI, and at the end of last year set a goal for AI to generate half of the platform’s code.

“As companies use multi-agent systems, employees find themselves switching between more tools,” the researchers wrote. “Contrary to the promise of having more time to focus on meaningful work, juggling and multitasking may become a defining feature of working with AI.”

AI incurs “significant costs” but can improve combustion

This AI-induced stress “results in significant costs in the form of increased employee errors, decision fatigue, and intentions to quit,” the researchers said.

Study respondents who said they had a brainstorming session experienced 33% more decision fatigue than those who didn’t, which researchers say can cost large companies millions of dollars a year. Those with AI brains are also about 40% more likely to actively intend to quit.

Those reporting AI brains also self-reported making serious errors about 40% more often than those who didn’t, with a serious error defined as having “serious consequences, such as those that could affect safety, outcomes, or critical decisions.”

However, researchers have found that using AI to replace repetitive and routine tasks reduces chronic stress in the workplace, which leads to negative feelings about work and reduced productivity.

related to: Anthropic reopens Pentagon talks as tech groups push Trump to drop danger label

Respondents who used AI to reduce time spent on routine and repetitive tasks reported 15% lower levels of fatigue than those who did not use AI in this way.

The researchers said that leaders of companies that want to reduce the AI ​​brain should “clearly define the purpose of AI in the organization” and explain how the workload will change with the tool.