Workers Who Direct AI Agents Outperform Peers
A joint study by the McCombs School of Business at the University of Texas at Austin and KPMG found that early-career workers who actively direct artificial intelligence systems outperformed those who simply delegate tasks to them. The research involved over 523 junior professionals at KPMG, revealing that individuals who treated AI as a collaborator—offering direction and oversight—produced significantly better results than peers with similar knowledge and skills. The study emphasized the importance of how employees interacted with AI rather than what they knew. Top performers were able to enhance AI's output by providing meaningful critiques and guidance, while underperformers often steered AI in unproductive directions or focused on irrelevant issues. Researchers recommended that organizations rethink employee value within AI systems and develop training programs to coach workers on effectively using these tools. The findings highlight a growing need for transparency in how employees use AI at work. By documenting decisions made around AI output, companies can identify skill gaps and improve workforce performance across all levels.
Topics: artificial intelligence employee performance HR research