AI-PACE: a framework for integrating AI into medical education

Area of Science:

• Medical Education
• Artificial Intelligence in Medicine
• Health Professions Education

Background:

• Medical AI education is fragmented and lacks structure, especially for generalist physicians.
• Current AI training is often short-term, biased towards procedural fields, and neglects the Affective domain.
• There is a need for a structured, longitudinal approach to AI education in medicine.

Purpose of the Study:

• To identify structural gaps in current medical AI education.
• To propose a novel framework for organizing AI competencies in medical education.
• To provide a structured approach to AI training for all physician levels.

Main Methods:

• An integrative review of 23 peer-reviewed articles published between 2016 and 2025.
• Analysis of identified structural gaps in AI education.
• Development of a framework based on Bloom's Taxonomy.

Main Results:

• Identified three key structural gaps: lack of reinforcement, procedural-field bias, and under-representation of the Affective domain.
• Proposed the AI-PACE (Psychomotor, Affective, Cognitive, Embedded) framework.
• The framework organizes AI competencies longitudinally across undergraduate, graduate, and continuing medical education.

Conclusions:

• The AI-PACE framework addresses identified gaps in medical AI education.
• It offers a structured, comprehensive approach to developing AI competencies for physicians.
• Implementation of AI-PACE can enhance AI literacy and application in clinical practice.