Back to News Inside the Research

Kines researchers use ChatGPT to adapt home fitness program for neurodivergent kids

Movement science faculty members Rebecca Hasson and Haylie Miller turned to the generative AI tool to modify instructions for at-home exercise videos so they better fit the needs of neurodivergent children.
A child in a navy blue polo and khaki pants stretches out his arms, grinning as children around him jump and exercise in a classroom.

A University of Michigan research team has used AI to help deliver exercise “snacks” to kids with autism.

The research team, led by U-M exercise physiologist Rebecca Hasson, overhauled a program that aimed to help children exercise when families had to pivot to online learning during the COVID-19 pandemic. The researchers from the School of Kinesiology used ChatGPT to tweak the instructions for the program’s 132 exercise videos, adapting them to better meet the needs of neurodivergent children.

For more, read the full story on Michigan News.