Environment Scan of Generative AI Infrastructure for Clinical and Translational Science

Area of Science:

• Healthcare Informatics
• Artificial Intelligence in Medicine
• Translational Science

Background:

• Generative AI (GenAI) and large language models (LLMs) present significant opportunities and challenges for healthcare.
• The Clinical and Translational Science Award (CTSA) network comprises 36 institutions actively exploring GenAI integration.

Purpose of the Study:

• To conduct an environmental scan of GenAI infrastructure within the national CTSA network.
• To assess institutional readiness, stakeholder roles, governance, and ethical considerations for GenAI adoption in healthcare.

Main Methods:

• A survey was administered to leaders of academic medical centers and health systems within the CTSA network.
• The survey focused on current GenAI deployment status, governance models, and identified challenges.

Main Results:

• Most institutions are in the experimental phase of GenAI deployment, exhibiting diverse strategies.
• Centralized decision-making is preferred, but significant gaps exist in workforce training and ethical oversight.
• Concerns regarding GenAI bias, data security, and stakeholder trust were identified.

Conclusions:

• GenAI integration in healthcare requires a coordinated approach involving senior leaders, clinicians, IT staff, and researchers.
• Addressing ethical concerns and ensuring robust governance are crucial for the effective implementation of GenAI.
• This study provides insights and a roadmap for healthcare institutions leveraging GenAI to enhance care quality and operational efficiency.