This research examines how LLMs may encode social biases that create inequities when deployed in robotic caregiving systems.

• Demographic variables significantly affect how LLM-powered robots provide care
• Robots may perpetuate societal biases in healthcare delivery based on patient characteristics
• Research identifies critical ethical considerations for AI deployment in vulnerable care settings
• Results highlight the need for bias mitigation strategies in medical AI systems

This work has profound implications for the healthcare industry as robotic caregivers become more prevalent, potentially introducing or amplifying disparities in care quality across different demographic groups.

Encoding Inequity: Examining Demographic Bias in LLM-Driven Robot Caregiving