Founder Reports surveyed 2,052 U.S. employed adults who use AI at work at least once a week to measure the time they spend botsitting. That’s the work of training AI, overseeing it, checking its output, and fixing what it gets wrong.

---

What’s Inside the Report

Workplace AI is usually judged by the time it saves. The time it demands in return gets far less attention. This report addresses that hidden workload and finds that AI isn’t paying off evenly. Heavy users have built their workflows around it and defend it strongly. Occasional users describe a tool that adds about as much work as it saves.

Download the free report for full details, including:

• 82.5% of workers who use AI weekly spend some time each week overseeing, checking, or fixing it, and 15.8% spend 6 or more hours.
• 53.2% of workers say they’d get the same amount of work done or more if their AI tools disappeared tomorrow. Only 13.6% say they’d get much less done. Even among workers who use AI multiple times a day, 37.6% say they’d do as well or better without it.
• If their main AI tool were an employee, 17.1% of workers would promote it and 16.5% would give it a warning or fire it.
• 28.5% of workers who use AI multiple times a day would promote it, compared to just 8.0% of weekly users. Among workers spending more than 10 hours a week on oversight, 32.9% would promote their AI and 41.0% say they’d get much less done without it.
• Among workers who’ve trained an entry-level employee, 60.0% say getting good work from AI takes at least as much effort, and 33.1% say it takes more.
• 61.8% of workers who spend any time managing AI say their employer doesn’t fully account for it, or they aren’t sure. For 34.5%, it isn’t counted at all.
• Light users are the most likely to go unrecognized. 53.4% of workers who spend less than an hour a week on AI oversight say that time goes uncounted, compared to 13.3% of those spending more than 10 hours.
• 40.7% of managers spend 3 or more hours a week on AI oversight, compared to 25.0% of individual contributors.
• Nonprofit workers would discipline their AI more than three times as often as they’d promote it (22.8% vs. 7.3%). At for-profit companies, the split is 18.7% promote and 16.0% discipline.
• 59.0% of women say they’d get as much done without AI, compared to 47.4% of men. The difference holds among multi-daily users (44.3% vs. 31.8%).

About the Survey

Founder Reports surveyed 2,052 employed adults in the United States in October 2026, all of whom use AI at work at least once a week. The survey carried a neutral title with no reference to oversight or productivity, so respondents weren’t primed toward any of the study’s themes before answering. Self-employed workers were excluded because several questions refer to an employer or a workload set by someone else.

The full report includes the complete methodology, sample composition, and breakdowns by AI usage frequency, hours of oversight, work role, employer type, sex, and age, along with charts for each finding.

The report is free to download. If you cite any of the data, we ask that you link back to this page.

For media inquiries, interview requests, or questions about the data, contact marc@founderreports.com.

Research by Founder Reports (founderreports.com)

The photo at the top of this page is by YuriArcursPeopleImages / Envato.