The Rise of the 'Botsitters'

Key Points:

• A report from Glean's Work AI Institute reveals that white-collar workers spend an average of 6.4 hours per week "botsitting" AI, involving tasks like feeding context, checking outputs, debugging, and cleaning errors, which consumes much of their time.
• Despite 87% of surveyed workers using AI and 75% feeling more productive, only 13% reported significant organizational performance improvements, highlighting a productivity paradox linked to unrecognized and unrewarded botsitting work.
• The burden of botsitting negatively impacts employee morale, with those spending more time on such tasks being 73% more likely to seek new jobs due to exhaustion and resentment from the extra, often tedious work.
• Workers often act as intermediaries between disconnected AI systems, fixing mistakes and providing missing context, while some are asked to automate parts of their jobs they find meaningful, risking job satisfaction and engagement.
• Organizations achieving better AI results invest in supporting employees with context, training, and clear standards for AI use, rather than just increasing AI deployment, aiming to reduce botsitting and retain talent.