Will Patients Accept Generative AI Genetic Counseling?

Poster Number: P131
Presentation Time: 05:00 PM - 06:30 PM

Abstract Keywords: Large Language Models (LLMs), Delivering Health Information and Knowledge to the Public, Patient Engagement and Preferences, Qualitative Methods
Primary Track: Applications
Programmatic Theme: Consumer Health Informatics

This study explores acceptance of AI-generated answers to genetic counseling questions. Patients evaluated human and LLM-generated responses to questions regarding Alzheimer’s risk. LLM responses were rated more relevant, trustworthy, and coherent than those from counselors. While patients seem receptive to LLM-provided information, their preferences for communication style vary, highlighting the need to tailor AI responses. This underscores the potential of LLMs in genetic counseling, warranting further research to prevent inaccuracies and address individual patient needs.

Speaker(s):
Yidi Huang, MS
University of Pennsylvania

Author(s):
Hita Kambhamettu, BS - University of Pennsylvania; Angela Bradbury, MD - Hospital of the University of Pennsylvania; Kevin Johnson, MD, MS - University of Pennsylvania;