AI frees up time at work — but does that time go into better work, or just more work piled on?
Do employees spend freed AI time on better work or just more tasks?
This explores what actually happens to the time AI is supposed to save at work: whether people put it into deeper, higher-quality work, or whether it gets absorbed by more tasks, faster pace, and cleanup.
This explores where AI's promised time savings actually go once they reach a real workplace: into better work, or just into more work. The short answer from the corpus is that the time mostly doesn't arrive as free time at all. It gets spent before anyone decides what to do with it. One line of research finds that AI doesn't shrink total task time so much as move it. Hours shift away from doing the work and toward writing prompts, reading outputs, and judging whether they're right Does AI really save time, or just change how we spend it?. That means 'time saved' is often a poor measure, because the work has changed shape rather than gotten shorter.
When researchers follow the hours, cleanup is the first thing that absorbs them. A Workday survey found that almost 40% of reported savings disappear into fixing errors and checking outputs, and only 14% of employees consistently come out ahead Where does AI's time savings actually go in practice?. Zapier's numbers look similar: 92% of users say they're more productive, yet they spend about 4.5 hours a week fixing AI's mistakes, and the heaviest users spend the most How much time do workers really spend fixing AI mistakes?. The cost also spreads to colleagues. Workers estimate that about 15% of the work they receive is unhelpful AI-generated content, and each instance takes nearly two hours to deal with. That's longer than it would have taken the sender to do the work properly How much work that employees receive is actually unhelpful AI content?. One person's saved hour can become a coworker's lost afternoon.
Whatever time is left over tends to go to 'more' rather than 'better.' A Berkeley Haas ethnography found that AI made people feel able to take on more. Their scope grew, natural stopping points disappeared, and work moved into what used to be breaks Does generative AI actually save workers time or intensify it?. ActivTrak's activity-tracking data points the same way: as AI adoption rose, people spent more time in work apps, worked more weekends, and had less daily focus time than at any point in three years Does AI adoption actually reduce the work that employees do?. Piling on tools makes it worse. BCG found self-reported productivity peaked at three AI tools and dropped with four or more, because keeping track of everything wore people down and raised their intent to quit Does using more AI tools always boost worker productivity?.
A less obvious point: some saved time may never be visible to the organization. In Anthropic's interview study, most workers said AI saved them time, but around 70% hid or downplayed using it because of stigma and worries about their jobs Why do workers hide productivity gains from AI use?. Executives' sense of the gains also runs ahead of what can be measured Do AI productivity gains feel larger than they actually measure?. So 'where did the time go?' is partly a question nobody can see the answer to.
The 'better work' outcome does show up, but only under certain conditions. Gains appear when people apply skills they already have, not when they use AI to learn new ones When does AI actually boost worker productivity?. They also appear when AI touches only a few of a job's tasks, which lets workers shift toward the parts AI doesn't touch Does concentrated AI exposure enable workers to adapt and reallocate?. The Workday data ties the minority who come out ahead to employers who retrained staff and redesigned roles, rather than just deploying tools Where does AI's time savings actually go in practice?. Acemoglu, Autor and Johnson explain why that redesign is rare: firms earn more from AI that automates expertise than from AI that creates new, better tasks for workers, so the 'better work' version gets too little investment Why do firms build automating AI instead of pro-worker AI?. 'Better work' isn't what happens by default. Someone has to design for it on purpose.
Sources 12 notes
Research shows AI doesn't reduce total task time; it reallocates it away from active work toward composing prompts and understanding outputs. This shift changes the cognitive demands and learning outcomes, making time-on-task a poor productivity metric.
A Workday-commissioned survey of 3,200 active AI users found that while 85% save 1–7 hours weekly, almost 40% of those savings disappear into correcting errors and verifying outputs. Only 14% of employees consistently see positive net outcomes, with success tied to organizations that retrain staff and redesign roles rather than simply deploying tools.
A Zapier survey of 1,100 enterprise AI users found 92% report productivity boosts, yet the average worker spends over half a day weekly revising AI-generated work. Trained, heavy users report the largest gains but also spend the most time on cleanup.
A September 2025 survey of 1,004 U.S. desk workers found respondents estimate 15.4% of work they receive is AI-generated but unhelpful content. Employees report spending an average of 1 hour 51 minutes dealing with each instance, longer than if the sender had done the work themselves.
A Berkeley Haas ethnography found AI didn't save time but instead expanded what workers felt capable of taking on, leading to faster pace, broader task scope, and work extending into former break times. Three mechanisms drove this: scope creep, dissolved stopping points, and multiplied parallel threads.
Show all 12 sources
ActivTrak's behavioral trace data show that as AI tool adoption rose sharply across monitored organizations, employees spent more time in work applications, more hours on weekends, and experienced a three-year low in daily focus time. The report concludes that AI amplifies the speed and density of work rather than reducing it.
BCG's survey found self-reported productivity rose with up to three AI tools but fell sharply with four or more. Workers experiencing this 'brain fry' showed 34% quit intent versus 25% without it, driven by oversight burden rather than tool count alone.
In a 1,250-person interview study, 86% of general workers and 97% of creatives said AI saved them time, yet 69–70% actively hid or downplayed their use due to workplace stigma and concerns about professional identity and economic displacement.
A survey of 750 executives found that perceived AI productivity gains exceed measured ones, likely because revenue lags operational improvements. Effects concentrate in high-skill services and finance, with labor reallocating rather than shrinking overall.
Studies showing AI productivity gains measured tasks within workers' existing domains. When workers used AI to learn new skills, productivity gains disappeared and learning suffered, suggesting prior findings do not generalize to skill acquisition.
Analysis of task-level AI exposure across firms 2010-2023 shows that while higher mean exposure reduces labor demand, more concentrated exposure (affecting few tasks) enables workers to reallocate to non-displaced tasks, producing modest net employment effects.
Acemoglu, Autor and Johnson argue that automating expertise generates higher economic returns for firms than creating new tasks, creating a collective-action gap where individual profit-maximization conflicts with worker welfare.
Papers this line draws on 8
The research behind the notes this line reads — ranked by how closely each paper relates.
- Beyond Productivity: Measuring the Real Value of AI
- Adoption of Generative AI in the Workplace: Increasing and Shifting the Balance of Productivity and Communication Activity
- 2026 State of the Workplace
- What 81,000 people told us about the economics of AI
- How AI Impacts Skill Formation
- Estimating AI productivity gains from Claude conversations
- Zapier Survey Finds Workers Spend 4.5 Hours Per Week Cleaning Up AI Mistakes
- Toward Measuring AI's Effects on Skill Formation: The Stock-Formation Gap