If you have ever worked alongside an automated system that misread your actions or responded in an unexpected way on the job, you have experienced the kind of breakdown University of Michigan (U-M) researchers are working to prevent. The project is focused on helping intelligent systems recognize when they misunderstand people and recover quickly to keep interactions safe, efficient and trustworthy.
The project, titled “Enhancing Worker Resilience in AI-Integrated Manufacturing: Exploring Error Recovery Strategies for Human-AI Interaction Failures,” is led by Manhua Wang, Assistant Professor in the U-M Department of Industrial and Operations Engineering (IOE). Wang is collaborating with Felix Warneken, Professor in the U-M Department of Psychology, to address a growing challenge in AI-enabled workplaces: failures stemming not from hardware or software issues but from misunderstandings between humans and intelligent systems.
The research team is currently developing its first study and anticipates beginning data collection during the summer of 2026.
Read the complete article at UM IOE
