Study Finds Workers Who Challenge AI Outperform Those Who Simply Accept Its Output

A new KPMG and University of Texas study of 523 professionals finds that workers who actively challenge and guide AI consistently outperform those who blindly accept its output, raising urgent concerns about AI overreliance deskilling the modern workforce.

• A new KPMG and University of Texas at Austin study of 523 early-career professionals reveals that workers who treat AI as a thinking partner, actively steering and challenging it, consistently outperform AI working alone.
• Three distinct user types emerge: 'AI amplifiers' who guide AI actively produce the best results, 'AI delegators' who accept AI output uncritically match only AI-level performance, and 'AI apprentices' who engage critically but poorly actually score below the AI baseline.
• Experts warn that as companies mandate AI use, overreliance is fueling deskilling and eroding critical thinking, making it essential for organizations to invest in training that strengthens the human skills needed to use AI effectively.