A new Glean Work AI Institute report found that white-collar workers spend an average of 6.4 hours per week “botsitting” AI systems, nearly a full working day.
The report, produced with researchers from Notre Dame, Stanford, and UC Berkeley, surveyed 6,000 full-time workers in the U.S., U.K., and Australia between December 2025 and January 2026. While 87% of workers said they use AI at work and 75% said it makes them more productive, only 13% said their organization was performing significantly better because of it.
The findings show a workplace challenge for companies adopting AI across departments.
Employees are spending time on AI supervision, including context input, output review, error correction, and work across disconnected systems. Rebecca Hinds, head of the Work AI Institute, described botsitting as “often tedious” and “exhausting” work that is “not rewarded and it’s not appreciated or tracked or measured and certainly not incentivized within the organization.”
Why It Matters: AI deployment alone does not guarantee productivity gains. The work required to make AI useful can become a hidden tax on employees when support systems are missing. Organizations that fail to address botsitting risk lower morale, weaker AI returns, and higher employee attrition among the people expected to make these systems function.
- AI Oversight Is Becoming a Measurable Workload: The report’s 6.4-hour weekly average shows that AI use is adding a regular supervision burden to the workday. Employees often have to guide the tools before the output is useful, then spend more time reviewing the results for accuracy. That extra labor matters because the time spent making AI usable can eat into the productivity gains workers expected.
- Productivity Gains Are Getting Lost Inside Daily Workflows: Many employees said AI helps them work more productively, yet far fewer said their organizations are performing significantly better because of it. The gap suggests that personal efficiency can fade when AI-assisted work still requires review and cleanup before it contributes to finished work, with 77% of workers reporting correcting or redoing AI-assisted work in the previous month. In that environment, AI may help with a task while leaving the organization with an added process burden.
- Disconnected Systems Are Making AI Feel Like Extra Work: Employees are supplying information that AI systems should already have access to and are having to carry work across tools that do not work well together, as 33% report using four or more AI tools a week. When that happens, workers become the link between systems rather than the beneficiaries of smoother workflows, which explains why employees can find AI useful while still feeling weighed down by the effort required to manage it.
- Botsitting Is Creating Retention Risk: Workers who spend an unusually large share of their AI time botsitting are 73% more likely to be actively looking for another job, according to the report. The risk grows when employees are expected to absorb AI cleanup work without recognition or support. As the report claims, “Workers who absorb it without recognition or reward grow exhausted. Then they grow resentful. Then they start polishing their résumés.”
- Better AI Outcomes Require More Structure Around Use: The report found that workers in organizations with no impact, negative impact, or unclear impact spend 49% of their AI-related work time directly inside AI tools, compared with 27% for workers in organizations reporting transformative impact. That difference suggests that better results depend on where AI fits into their work and how outputs are judged.
Go Deeper -> The rise of the ‘botsitters’ – Business Insider
Trusted insights for technology leaders
Our readers are CIOs, CTOs, and senior IT executives who rely on The National CIO Review for smart, curated takes on the trends shaping the enterprise, from GenAI to cybersecurity and beyond.
Subscribe to our 4x a week newsletter to keep up with the insights that matter.
