AI Tools Save 11 Hours Weekly but Require 6 Hours of 'Botsitting'

Artificial intelligence tools are delivering significant time savings to office workers, but those gains come with a substantial hidden cost: the hours spent supervising, correcting, and refining what the AI produces.

A survey of 6,000 digital workers across the United States, United Kingdom, and Australia found that AI saves each worker roughly 11 hours per week. However, those same workers spend more than six hours weekly on what researchers call "botsitting"—checking AI output, fixing mistakes, and rerunning prompts to get usable results.

The productivity paradox

The research, published by the Work AI Institute with contributors from Stanford University and UC Berkeley, reveals a striking disconnect. While 75% of individual workers report productivity gains from AI, only 13% of organizations say they've seen significant business benefits from AI adoption.

For every hour a worker spends getting useful output from AI, they spend roughly another hour making that output usable. Of total AI interaction time, 37% goes to botsitting while 36% goes to actual productive work.

"Most people don't realize the amount of time that they're spending working on the tools to get the time savings that they're professing," said Paul Leonardi, a co-author of the study and professor of technology management at UC Santa Barbara.

Why AI sessions fail

The time drain stems partly from how often AI tools simply don't deliver. More than a third of AI sessions fail outright, requiring workers to start over or substantially rework the output. Workers must gather the right files, documentation, and context to produce quality results—a process that consumes significant time.

The survey analyzed anonymized workplace data from companies using the Glean Work AI platform between December and January. The research found what it calls a "thick, mostly invisible layer of human labor holding the whole thing together."

The delegation problem

As workers hand over larger portions of their jobs to AI, a new risk emerges: 41% of workers say they sometimes deliver AI-generated work they couldn't explain if asked.

The report cites an example of a junior software engineer who pasted thousands of lines of AI-generated code before going to bed. When something broke, a senior engineer already behind on deadlines had to untangle the problem while the junior engineer struggled to explain the code.

"I think what's happening with a lot of these Gen AI tools right now is we're essentially expecting individual contributors to act as managers," Leonardi said. "They're just managing these AI tools, AI agents, and we're expecting that they'll be able to produce way more, but we're not taking into account all of the work that actually goes into managing."

Why it matters

This research challenges the narrative that AI automatically translates to business value. Companies pushing employees to maximize AI usage—as many Silicon Valley firms have done over the past six months—may be creating busywork rather than genuine efficiency gains. The gap between individual productivity and organizational results suggests that without better implementation strategies, AI tools risk becoming expensive distractions rather than transformative business assets.

These findings were first reported by the Los Angeles Times, with research conducted by the Work AI Institute, which is sponsored by AI company Glean.