The burnout from "botsitting" leads to "botshitting"—a slow surrender of agency where workers ship unverified AI outputs. This creates a vicious cycle of low-quality work, increased rework, and moral disengagement, with 40% of workers blaming AI for failures instead of themselves.
A new, invisible form of labor called "botsitting"—feeding context, checking outputs, and debugging—consumes 37% of workers' AI time. This is more time than they spend actively using AI to complete tasks (36%), creating a significant, hidden productivity drain and burnout risk.
Paradoxically, the AI tools users rate as most productive, like ChatGPT and Claude, are also linked to the highest rates of "botshitting" (shipping unverified work). This suggests that as AI becomes more capable, the risk of user over-reliance and declining quality control increases significantly.
Despite employees saving 11 hours weekly with AI, only 13% of organizations see significant improvement. This highlights a structural failure to translate individual efficiency into organizational effectiveness, a problem that exists even without the cost of "botsitting"—the hidden labor of managing AI.
High-achieving managers delegate 32% more coordination work to AI, not to replace management but to eliminate administrative tasks. This frees them up for high-value work like coaching, which in turn doubles their team's trust in using AI for sensitive decisions like performance reviews.
An employee is 5.6 times more likely to adopt AI if a cross-functional teammate uses it—a far greater influence than leaders (2.4x) or direct teammates (3.2x). This is because cross-functional users build tools that solve the messy, real-world coordination problems that plague organizations.