🔍 Read the full analysis: Can AI Simplify A Workday Spent Switching Between Apps? on ThorstenMeyerAI.com
Get smart everyday buys delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
A Harvard Business School preprint analyzed 103 million application events from 1,017 workers at eight Indian companies. It found an average of 908 app switches per workday; switching fell in the ten app uses after AI use but was higher overall on days when workers used AI. The observational data cannot show whether AI caused those patterns or improved work.
A Harvard Business School preprint tracking 1,017 office workers found they switched applications an average of 908 times per workday, but its findings do not show that AI simplifies the workday: workers switched less in the ten app uses after using AI, while making about 200 more switches on AI-use days overall. The study is observational, so it cannot establish whether AI changed workers’ habits or was used on days that were already more fragmented.
The paper, Digital Fragmentation and Generative AI Use Across 103 Million Application Events, was written by Sumer S. Vaid and Ashley V. Whillans of Harvard Business School. Researchers analyzed 103 million application events recorded from November 2025 through January 2026, covering a median of 58 workdays for each participant. The workers were employed by eight organizations in India across software, accounting, financial services, e-publishing, law and marketing.
Across 62,693 person-days, the researchers counted 56.9 million switches between applications. The average was 908.3 switches per day; the median employee made 814. The middle half of employees recorded between 547 and 1,130 switches per workday. An application remained active for a mean of 27.8 seconds before a switch, while the median interval was 3.3 seconds. These figures describe recorded computer activity, not the quality or output of the work.
The AI findings point in two directions. In the ten application uses after an AI use, employees used fewer apps, switched less, stayed about half a second longer in each app and moved between apps in a more predictable pattern. Yet on days when workers used AI, they made about 200 more switches per day; days with more AI minutes were also more fragmented. The paper does not establish why the two patterns coexist.
App switching and AI · one moment
What happens around one AI query?
The study zoomed in on every AI use: the ten app uses before it, and the ten after. Up close, work calms down a little. Zoom out to the whole day, and the picture changes.
- 1 · A workday of switchingWorkers changed apps 908 times a day on average. One app held their focus for 27.8 seconds on average, 3.3 seconds at the median.
- 2 · AI comes in at a busy momentPeople used AI a median of 13 times a day, for a median of 7 seconds. Fragmentation peaked in the hour of AI use.
- 3 · The next ten app usesFewer different apps, fewer switches, about half a second longer per app, more predictable. Real, but small.
- 4 · Zoom out: the whole dayOn days with AI use, the same people made about 200 more switches. Whether AI calms work or busy moments bring AI: unknown.
Source: Sumer S. Vaid, Ashley V. Whillans (Harvard Business School), “Digital Fragmentation and Generative AI Use Across 103 Million Application Events”, arXiv:2607.06681v2, preprint under review. Blocks are schematic, not to scale.
Graphic: Powered by Thorsten Meyer AI
Why AI Did Not Mean Fewer Daily Switches
The findings complicate the claim that adding AI automatically gives workers a calmer or more focused day. AI use was associated with a short-term pattern of fewer switches after an interaction, but also with more switching across the day. Those are different time windows, and neither result proves that AI caused the change.
For employers deciding whether to buy AI tools to reduce distraction, the distinction matters. This study did not test a tool against a control group or measure whether employees finished work faster, produced better results or felt less interrupted. It therefore offers no evidence that AI improved productivity or that an AI purchase would reduce app switching. It does provide a large-scale account of how recorded application use varied across workdays.
The paper also reports that day-to-day changes within the same worker accounted for 44.6% of variation in fragmentation, compared with 35.8% associated with differences between people and 19.6% between organizations. That pattern suggests that work conditions on a particular day may be as relevant as stable differences between workers, though the analysis does not identify the causes.
Products Worth Considering

Vari ComfortEdge 48×30 Inch Adjustable Electric Standing Desk

Logitech MX Brio Ultra HD 4K Webcam

Apple iMac MK472LL/A 27‑Inch Retina 5K Desktop (3.2 GHz Intel Core i5, 8 GB DDR3, 1 TB, macOS)
As an affiliate, we earn on qualifying purchases.
How the Workplace Activity Was Measured
The data came from Flowace, a commercial workforce-analytics service used by the participating organizations. Its software recorded when the active application changed, along with details such as app name and time. The paper says records were provided to researchers in anonymized form under a data-use agreement, and that Harvard’s human-subjects committee approved the research protocol.
The employers determined which applications counted as “productive,” according to the source material. The sample covered workers at eight firms in India, rather than a representative group of employees across countries or industries. The paper states that it received no external research funding and that Flowace provided no commercial support and had no role in study design, analysis or interpretation. One author disclosed equity in a separate private company that uses AI to create enterprise digital twins.
The paper first appeared on arXiv on July 7, 2026, and was revised on August 7. It identifies itself as a preprint under peer review and warns that its findings may change. A headline circulating on November 18 emphasized the lower switching immediately after AI use; the overall daily result is a separate part of the study and changes the interpretation.
“The manuscript is currently under review, and the findings may change based on the peer review process.”
— The study’s authors
What the Observational Data Cannot Show
The study cannot determine whether AI caused more daily app switching, whether employees turned to AI because their days were already busy, or whether both processes occurred. The authors describe those possibilities and do not treat the data as an experiment. The reported association between AI use and switching should not be read as evidence of cause and effect.
It is also unclear whether the patterns apply outside these eight Indian organizations. The source material says the analysis did not cover phone use and did not assess whether work quality or performance improved. App-switch counts are not a direct measure of interruption, attention or productivity, and employer-selected productivity labels may shape what the monitoring system records as work.
The figures on time spent in applications are descriptive averages and medians; they do not mean every worker changed apps at the same pace. Comparisons with earlier attention studies cited in the paper also involve different methods, so they are a broad comparison rather than a precise measurement of change over time.
Peer Review and Better Tests Ahead
The immediate next step is peer review, which could lead to revisions to the methods, analysis or conclusions. Until then, the paper should be treated as a preprint rather than settled evidence about AI and workplace focus.
To test whether AI reduces fragmentation, future research would need to separate the effect of AI from the kinds of days and tasks that prompt employees to use it. It would also need outcomes beyond app counts, such as task completion, work quality and employee experience. For now, the study’s authors do not establish that AI simplifies a workday, and the available data do not support buying an AI tool specifically as a focus aid.
Key Questions
Did the study find that AI reduced app switching?
It found fewer switches in the ten application uses after AI use, but workers made about 200 more switches on AI-use days overall. The results describe different time windows and do not establish that AI caused either pattern.
How many workers and application events did researchers study?
The Harvard Business School researchers analyzed 103 million application events from 1,017 office workers at eight organizations in India, between November 2025 and January 2026.
Was the study peer-reviewed?
No. The paper was posted as a preprint under peer review. Its authors state that findings may change during that process.
Does the research show AI improved productivity?
No. The study counted application activity and did not assess whether AI improved work quality, output or completion time. Its observational design also cannot establish that AI caused the switching patterns.
Can the results be applied to all workers?
That is not established. The sample came from eight organizations in India, and the source says the study did not track phone use. Results may differ in other countries, workplaces or work arrangements.
Source: ThorstenMeyerAI.com
