🔍 Read the full analysis: What AI Means For Workers Who Switch Apps Hundreds Of Times on ThorstenMeyerAI.com
Get business pricing on tech for your team
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
A Harvard Business School preprint analysing 103 million application events found workers switched apps an average of 908 times per day. Workers switched less in the ten app uses after AI use, but made about 200 more switches on days they used AI; the observational data cannot show that AI caused either pattern.
A Harvard Business School preprint analysing 103 million application events from 1,017 office workers found that employees switched apps an average of 908 times per day, while AI use showed two different associations: fewer switches in the ten app uses after an AI session, but about 200 more switches on AI-use days. The study is observational, so it does not establish that AI caused either change.
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. It uses activity records covering 62,693 person-days from November 2025 through January 2026. Researchers received anonymised data from Flowace, a commercial workforce-analytics service used by the participating employers.
Across the records, workers made a mean of 908.3 switches per day; the median employee made 814. The middle half of employees ranged from 547 to 1,130 switches per workday. The researchers define a switch as a change to a different application, rather than a move between two documents within the same app. An app remained active for a mean of 27.8 seconds before the next switch, with a median of 3.3 seconds.
The AI findings depend on the time window. In the ten application uses after an AI use, employees used fewer apps, switched less often, stayed around half a second longer per app and followed more predictable sequences. But on days when workers used AI, they made about 200 additional switches per day; days with more AI minutes were also more fragmented. These are associations in workplace logs, not results from an experiment.
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
AI Use and the Shape of Workdays
The findings complicate the idea that AI necessarily reduces the interruptions workers face. A short period of less switching after an AI interaction can coexist with a more fragmented day overall. For employees, managers and organisations considering AI tools, the distinction matters: an immediate pattern around an AI session does not show that using AI improves concentration across a whole workday.
The paper also points to the workday itself as a potentially important source of variation. The authors attribute 44.6% of observed fragmentation differences to day-to-day changes within the same person, compared with 35.8% to differences between people and 19.6% to differences between organisations. That suggests work schedules and daily demands may matter alongside individual habits, although the study does not identify which conditions cause more switching.
The evidence does not establish whether the additional switching reflects AI tasks requiring workers to move among tools, a busy day prompting workers to use AI, or a combination. It also does not measure whether those switches harm productivity or work quality. The results are not a basis on their own for treating AI as a focus aid or prescribing a particular workplace policy.
Products Worth Considering

Logitech MX Mechanical Mini for Mac Wireless Keyboard, Space Gray

LG UltraGear 27-inch QHD OLED Gaming Monitor, 240Hz, 0.03ms, G-SYNC Compatible, FreeSync Premium Pro, HDR True Black 400

Samsung 27-Inch Odyssey G5 QHD Gaming Monitor (G50F)
As an affiliate, we earn on qualifying purchases.
How the Workplace Data Was Collected
The sample included eight organisations in India, spanning software, accounting, financial services, e-publishing, law and marketing. Each person contributed a median of 58 workdays. Flowace software recorded when the active application changed, along with app names and other activity details. The employers decided which applications counted as “productive.” The researchers say the data was anonymised before they received it and that a human-subjects committee approved the study protocol.
The paper was posted to arXiv on July 7, 2026, and revised on August 7. It describes itself as a preprint under review and cautions that findings may change during peer review. The authors report no external research funding; they say Flowace provided no funding or commercial support and had no role in the study design, analysis or interpretation. The paper also discloses that Vaid holds equity in an unrelated private company that uses AI to create enterprise digital twins.
The reported 908 switches is a daily average from this sample and measurement system, not a universal count for all workers. The study did not cover other countries or phone use, and its app-activity records do not show whether employees completed tasks well or how they experienced the monitoring.
“The manuscript is currently under review, and the findings may change based on the peer review process.”
— Vaid and Whillans, in the preprint
What the Switching Data Cannot Show
The study cannot establish whether AI use reduces switching in the moments afterward, increases it across the day, or does neither. Workers were not randomly assigned to use AI, and the researchers say the direction of the relationship is unknown: AI may reorganise a workday, workers may reach for AI when their day is already consolidating, or both patterns may occur.
It is also unclear how well these results apply beyond the participating Indian companies, the monitored computer applications and the November 2025-to-January 2026 period. The data do not measure phone activity, task completion, output quality or worker wellbeing. Because monitoring software supplied the records and employers classified apps as productive, the measures reflect that system’s definitions as well as employees’ activity.
The headline figure of 908 switches should not be read as evidence that each worker experienced 908 harmful interruptions. The paper counts application changes, not the purpose of each change or its effect on concentration. Peer review may also lead to changes in how the findings are presented or interpreted.
Peer Review and Workplace Evidence
The paper’s findings remain subject to peer review, and the authors warn that they may change. Further research would need to test the relationship between AI and app switching under designs that can distinguish cause from correlation, while measuring work outcomes such as task completion and quality.
For now, the study supports caution rather than a firm prescription. Employers evaluating AI tools can distinguish short-term activity patterns from day-level effects and avoid treating fewer switches immediately after AI use as proof of improved focus. Whether the observed patterns recur in other countries, organisations and kinds of work remains to be tested.
Key Questions
How many times a day did workers switch apps?
The study found a mean of 908.3 application switches per day across its sample. The median employee made 814 per workday. The figure counts changes between different applications, not switches between documents inside one application.
Did AI reduce app switching?
The results varied by time window. Workers switched less in the ten application uses after an AI use, but made about 200 more switches on days when they used AI. The observational study cannot determine whether AI caused either pattern.
Who took part in the study?
The analysis covered 1,017 employees at eight organisations in India, across software, accounting, financial services, e-publishing, law and marketing. Activity records were collected from November 2025 to January 2026 through Flowace monitoring software.
Has the study been peer-reviewed?
No. The paper was posted as a preprint and says it is under review. Its authors caution that the findings may change during the peer-review process.
Does the study show that AI improves productivity or focus?
No. It measured application activity and switching, not productivity, work quality or worker wellbeing. Its results show associations, not that AI improves focus or causes more fragmented days.
Source: ThorstenMeyerAI.com
