Texas Christian University Professor Will Drover and co-author Laura Huang, a professor at Northeastern University, reveal why artificial intelligence (AI) adoption advances unevenly across tasks and roles, and what leaders should know to navigate the transformation.
December 10, 2025
By TCU Neeley School of Business
Will Drover, associate professor and department chair of Entrepreneurship and Innovation at the Texas Christian University’s Neeley School of Business, cuts through the noise surrounding artificial intelligence (AI) and offers business leaders a more grounded, human-centered view of how AI is truly unfolding. While headlines predict rapid, sweeping automation, his new article in MIT Sloan Management Review – a top global outlet at the intersection of research and practice – with coverage also appearing in The Wall Street Journal, explains that AI’s takeover is far more uneven and task-specific than many assume.
His framework identifies the 'friction factors' that determine whether AI will assist, reshape, or fully replace work in different domains – helping to explain why automation races forward in some areas while remaining stubbornly human for decades, if not forever, in others. The work gives leaders a more nuanced strategic lens to navigate workforce transformation.
The research, developed by Drover and co-author Laura Huang, a professor at Northeastern University, identifies ten specific friction factors that determine why AI adoption varies so dramatically across roles and industries. These include repetition, error tolerance, judgment, human assurance, relational depth, and organizational inertia – each of which helps explain why some tasks automate quickly while others still depend on distinctly human skills and oversight. When these dials are set to low friction, AI floods in wholesale; when dialed to high friction, AI slows or stalls, explaining the nuanced pace we're seeing in practice. When leaders plan based solely on job categories rather than task-level demands, they risk misjudging AI's pace and misallocating critical resources.
An example from the research is the continued use of two pilots in commercial airline cockpits. While aircraft technology has automated many functions, a bundle of frictions such as regulatory oversight and expert human judgement, keep full pilot automation far from reach. Automation continues, but not at the pace many predict and it still hasn’t eliminated the need for a human element.
“The barrier to full automation isn’t raw capability. It’s a stack of human, legal, and cultural constraints,” according to the research findings. Yet illustrating variation even within an industry, cargo and military aviation face far lower friction; lighter regulation, reduced human assurance demands and marginally higher error tolerance mean autonomous freight and uncrewed fighter jets are advancing much faster toward full automation. It's a clear demonstration that AI's trajectory is situational, not uniform, and determined by the specific friction factors at play. The research also unpacks AI’s nuanced pace in other domains, such as self-driving cars, elementary teachers, and medicine.
Drover’s analysis highlights a crucial leadership skill: anticipation. Leaders can better predict where automation will accelerate and where it will stall by breaking work into tasks and assessing key friction factors in play. This perspective reshapes workforce planning and strategic investment, moving the conversation from fear of replacement toward informed, targeted transformation as roles evolve.
For business leaders navigating AI transformation, Drover’s message offers both clarity and direction. Understanding the automation arc and the factors that slow or accelerate automation enables organizations to invest strategically, plan realistically, and lead teams with confidence. The future of work won't be defined by AI's capabilities alone; it will be shaped by the friction it meets, task by task and role by role.
