Harvard Study Suggests AI Expands Workloads

- A study by researchers from UC Berkeley's Haas School of Business, Aruna Ranganathan and Xingqi Maggie Ye, found that at a 200-employee tech company, AI intensified workloads through task expansion, blurring of work-life boundaries, and increased multitasking. One engineer in the study noted, "You had thought that maybe, oh, because you could be more productive with AI, then you save some time, you can work less. But then really, you don't work less. You just work the same amount or even more." - The study identified a cycle where AI accelerates tasks, which in turn raises expectations for speed and responsiveness, leading to a greater reliance on AI and an expansion of the quantity and density of work. This phenomenon is described as a continual switching of attention between tasks and checking AI outputs, which increases cognitive load. - A separate study by Harvard Business School and Boston Consulting Group involving 758 consultants introduced the concept of the "jagged technological frontier," where AI significantly boosts productivity for tasks within its capabilities. For these tasks, consultants completed 12.2% more tasks, were 25.1% faster, and produced 40% higher quality results. - The same HBS/BCG study found that for tasks outside of the AI's capabilities, consultants using AI were 19 percentage points less likely to produce correct solutions, highlighting the risks of misapplying AI. This has led to the concept of an "AI verification tax," where employees spend additional time checking and correcting AI outputs. - Wharton Professor Ethan Mollick, a researcher on the HBS/BCG study, describes two successful models for human-AI collaboration to navigate this "jagged frontier": the "Centaur" model, where there is a clear division of tasks between human and AI, and the "Cyborg" model, which involves a deep and constant integration of AI into the workflow. - While some studies indicate an increase in workload, other research suggests significant time savings. An LSE study of nearly 3,000 workers found that AI users save an average of 7.5 hours per week. Similarly, a Federal Reserve Bank of St. Louis study found that 33.5% of daily AI users reported saving four or more hours per week. - To mitigate the burnout associated with increased workloads, some companies are using AI to analyze work patterns, optimize task allocation, and send employees reminders to take breaks. There's also a push for establishing clear "AI practices" – a set of norms and procedures to govern the use of AI tools.