AI Productivity Has 'Intense' Downside, Study Finds

Got story updates? Submit your updates here. ›

A new study from UC Berkeley researchers Aruna Ranganathan and Xingqi Maggie Ye found that employees who enthusiastically adopted generative AI tools at a tech company worked at a faster pace, took on a broader scope of tasks, and worked longer hours - often without being asked. However, this surge in productivity proved unsustainable, leading to burnout, cognitive fatigue, and weakened decision-making. The researchers observed employees blurring the lines between work and non-work, taking on more multitasking, and expanding their responsibilities beyond their normal roles.

Why it matters

As AI evangelists like Elon Musk and Bill Gates promise a future with less work, this study suggests the reality may be more complex. While AI can boost productivity in the short term, the researchers warn that the changes it brings can be 'unsustainable, causing problems down the line' for employees. This highlights the need for employers to carefully manage the transition to generative AI tools to protect worker wellbeing.

The details

Over an 8-month period, the researchers conducted in-person observations, monitored internal communications, and interviewed 40 employees across engineering, product, design, research, and operations. They found that generative AI intensified work in three main ways: task expansion (employees taking on more responsibilities), blurred work/non-work boundaries (squeezing work into breaks), and increased multitasking (managing multiple active tasks simultaneously). This led to burnout, fatigue, and lower-quality work.

• The study was conducted from April to December 2025.

What they’re saying

What’s next

The researchers suggest that employers should implement more intentional pauses or dedicated break periods for employees, as well as sequencing work to avoid constant activity and 'siloing' of employees from the rest of the company.