• Eye Opening
  • Well Structured
  • Concrete Examples


Organizational ethnographer Matt Beane posits that artificial intelligence compromises learning for students in advanced fields. Traditional on-the-job learning involves observing an expert at work, perfecting the skills observed through practice and imparting those skills to others. However, as robots perform more of the skilled aspects of specialized work, students struggle to access the hands-on experience they need. Educators in all fields will appreciate Beane’s suggestion to refocus AI to support on-the-job learning. 


Resident surgeons learn their trade by watching attending surgeons perform operations. The residents gradually gain hands-on experience by struggling with progressively more complex tasks under guidance until they can perform surgeries independently and impart those skills to more junior doctors. Whether you call it “apprenticeship, coaching, mentorship” or “on-the-job training,” learning at work follows a standard protocol: “See one, do one, teach one.” Such learning is common to many professions, and this model has formed the backbone of on-the-job learning for millennia. However, the introduction of ...

About the Speaker

Matt Beane is an assistant professor of technology management at the University of California, Santa Barbara.

More on this topic

Customers who read this summary also read

“Human-compatible AI” by Stuart Russell (Data Driven Paris)
The Danger of AI Is Weirder than You Think
Work: In Progress
Meet the Leader with Jack Ma
The World in 2019
How Do We Find Dignity at Work?

Related Channels