Workflows built around multiple AI agents and constant tool switching are adding cognitive strain across large enterprises. A recent Harvard Business Review analysis describes this pattern as “AI brain fry,” a form of mental fatigue tied to intensive use and oversight of AI systems.
Employees increasingly manage clusters of agents that generate code, synthesize information, and produce drafts at high speed. Performance systems in some organizations reward activity metrics such as token consumption and AI output volume. This structure pushes workers to monitor more systems and outcomes within the same workday.
Employee-reported Al brain fry, by industry role (Source: Harvard Business Review)
Researchers studying workforce and AI trends surveyed full time U.S. employees at large companies across industries, roles, and seniority levels. Respondents detailed patterns of AI use, work experiences, and cognitive and emotional conditions tied to those workflows.
Participants described a “buzzing” sensation, mental fog, difficulty focusing, slower decision making, and headaches after extended AI oversight. Researchers define AI brain fry as mental fatigue from excessive use or oversight of AI tools beyond a person’s cognitive capacity.
Roles requiring sustained monitoring of AI systems demanded greater mental effort and produced higher levels of fatigue and information overload. Employees who said AI tools increased their workload also reported heavier cognitive strain. Oversight demands combined with added responsibilities expanded the number of outcomes employees had to track during the workday.
The number of tools used at the same time also influenced outcomes. Using a small set of AI tools aligned with productivity gains. Adding more tools reduced those gains. The pattern reflects limits on multitasking and attention.
Prevalence varies by role
Researchers asked participants whether they had experienced mental fatigue tied to intensive AI use. A share of workers who rely on AI reported that experience.
Legal roles reported the lowest prevalence. Marketing roles reported the highest. People operations, operations, engineering, finance, and IT also ranked high relative to other functions.
Participants frequently used terms such as “fog” and “buzzing.” They described extended back and forth with AI tools followed by difficulty thinking clearly, slower decisions, and the need to step away from screens to reset.
One senior engineering manager described juggling multiple tools used to weigh technical decisions, generate drafts, and summarize information. Constant switching and verification created a sense of mental clutter. Effort shifted from solving the core problem to managing the tools.
A finance director described extended AI assisted drafting and synthesis that left them unable to judge whether the output made sense. Work paused until the next day to recover focus.
These accounts reflect patterns of information overload and task switching associated with cognitive strain. Intensive AI oversight adds to that burden.
Measurable business costs
Cognitive strain linked to AI brain fry carries operational effects. Workers who reported the condition also reported higher levels of decision fatigue, meaning fewer mental resources available for high quality decisions.
Error frequency increased. Participants experiencing AI brain fry reported more minor mistakes that are easy to correct and more major mistakes with consequences for safety, outcomes, or significant decisions.
Retention signals also shifted. Workers reporting AI brain fry were more likely to express intent to leave their jobs. Many employees using AI most intensively fall into high performance talent pools that organizations seek to retain.
Where AI reduces strain
AI use varied by task, and workers who used it to reduce time spent on routine and repetitive work reported lower burnout levels. Handing off repetitive duties created more room for creative work, collaboration, and higher value tasks.
Team support and leadership influenced mental fatigue levels. Employees reported less mental fatigue when managers made time to answer questions about AI. Teams that integrated AI into shared workflows experienced less strain than those where individuals adopted tools independently.
Organizational signals shaped the experience. Employees who believed their companies expected more output because of AI reported greater fatigue. Employees who felt their organizations valued work life balance reported less strain. Guidance on how AI fits into daily work reduced cognitive pressure across teams.
“Organizations should evolve people analytics measures to monitor overall cognitive load and safeguard against mental fatigue linked to AI use as a novel job-related risk,” researchers concluded.
