From Agents to Governance: Essential AI Skills for Clinicians in the Large Language Model Era

Area of Science:

• Medical Informatics
• Artificial Intelligence in Medicine
• Clinical Practice

Background:

• Large language models (LLMs) are increasingly integrated into clinical settings.
• There is an urgent need for clinicians to develop competencies for responsible AI use.
• Balancing AI opportunities with accountability is crucial for patient safety.

Purpose of the Study:

• To propose a 3-tier competency framework for clinicians using AI.
• To guide the evolution of clinicians from AI users to responsible AI stewards.
• To enhance safe and effective AI integration in healthcare.

Main Methods:

• Development of a 3-tier competency framework: Foundational, Intermediate, and Advanced skills.
• Tier 1 focuses on safe use: prompt engineering, human-AI interaction, privacy, and consent.
• Tier 2 emphasizes evaluation: bias detection, explainability, and workflow integration.
• Tier 3 addresses leadership: ethical governance, regulatory strategy, and model management.

Main Results:

• The framework outlines specific competencies for each tier, progressing from basic to advanced AI stewardship.
• Competencies include prompt engineering, bias detection, ethical governance, and model life cycle management.
• Integration into medical education and job descriptions is proposed.

Conclusions:

• The proposed framework supports clinicians in safely and responsibly using AI tools.
• Standardized AI deployment through this framework can enhance clinical practice.
• Effective AI integration has the potential to improve patient outcomes and safety.