US desk workers spend more AI time fixing answers than doing the work
Reviewed by Helen Jones
BambooHR published Redesigning Work: AI's Performance Review on 1 September 2026, based on a survey of 1,608 full-time salaried US desk workers, including 520 HR managers and above. People reported 87 minutes of daily AI use, about 47 eight-hour workdays a year. Of that time, 42 percent went to troubleshooting errors and rewriting prompts, 35 percent to work they judged productive, and 23 percent to exploring. The troubleshooting share is about 36 minutes a day, or roughly 20 workdays a year. CFO.com later noted executives averaged 101 minutes a day against 54 for others. Sixty-three percent of organisations had raised AI tool budgets. Sixty-five percent of workers still felt confident and enthusiastic, and 58 percent said time savings was their main reason for using AI. The figures are self-reported, not a controlled productivity study.
Until now, the usual story was that more AI minutes meant more output. Teams treated blank-chat prompting as the job, then blamed the model when the first draft needed a rewrite. Budget increases looked like proof of progress because usage was easy to count and quality was not. This survey splits the clock. Most of the time people call AI work is still briefing, iterating, and checking. Enthusiasm and spend can rise while the share of immediately useful output stays the smaller slice. That is a workflow problem, not a motivation problem: if the same context, standards, and review steps are restated every session, the 20 lost days are the cost of having no reusable system.
Analysis
Treat this as a trap to avoid, not a reason to cut AI. Pick one recurring task you already run through a chat, log for five working days how much of that time is first-pass output versus restating context and repairing the draft, then lock the briefing, examples, and quality check into a Project, custom GPT, or shared template so the next person does not start from a blank box.
Source note
Pulse published by Collab365 Spaces, reviewed by Helen Jones on . Cite as "US desk workers spend more AI time fixing answers than doing the work", Collab365 Spaces. 2 sources referenced.