TED Talk: UCSB Asst Prof Matt Beane -- Impact of Artificial Intelligence on Training and Development

"Shadow Learning" the Result of Limited Opportunities for Hands-on Training and Mentorship

University of California Santa Barbara (UCSB) Assistant Professor Matt Beane presents insights into his research addressing the impact of artificial intelligence on the workforce in a newly released TED Talk available on the Ted.com homepage. Based on a three-and-a-half-year research study conducted while a graduate student at the Massachusetts Institute of Technology (MIT) and a faculty member at UCSB, Beane’s findings show that current deployment of artificial intelligence (AI), or intelligent machines, is not only inhibiting critical on-the-job learning but potentially creating obstacles to AI-driven productivity gains. In rare cases, according to Beane, workers are creating so-called “deviant” means to acquire the capabilities they need – a phenomenon he has defined as “Shadow Learning.” In his research, “Shadow Learning: Building Robotic Surgical Skill When Approved Means Fail,” published in Administrative Science Quarterly in 2018, Beane finds that in our quest for AI-driven productivity, we are compromising critical, presupposed pathways for learning on the job.

“Intelligent machines offer unprecedented efficiency and quality improvements. And while many of us may lose or gain jobs as a result, many, many more of us will have to adapt to these technologies in the jobs we have,” said Beane. “Unfortunately, we’ve been redesigning work to take advantage of these technologies far faster than we’ve been redesigning learning and development. Ironically, these intelligent machines are at the center of the trouble. We’re currently deploying AI in ways that are actually preventing workers from learning by doing – the most common and effective process for getting the new skills to adapt to these new technologies.”

Read the full press release here. 

Photo courtesy of TED.com.

Share this with FacebookShare this with TwitterShare this with LinkedInShare this with RedditShare this with EmailPrint this