Compounding advantage: Medical AI and structural injustice

Area of Science:

• * Medical Ethics
• * Health Equity
• * Artificial Intelligence in Medicine

Background:

• * Artificial intelligence (AI) in medicine is often framed as a democratizing force.
• * Prevailing market governance structures may lead to AI exacerbating existing institutional advantages.

Purpose of the Study:

• * To analyze how AI in clinical medicine functions under market governance.
• * To evaluate AI's impact on healthcare equity and identify the nature of injustice it produces.
• * To apply theoretical frameworks of social injustice to the medical AI context.

Main Methods:

• * Critical analysis of AI's role in clinical medicine within market-driven systems.
• * Application of Iris Marion Young's theory of structural injustice.
• * Examination of distributive justice principles (Rawlsian, Daniels) and their limitations.

Main Results:

• * AI acts as a self-amplifying mechanism, concentrating advantages and deepening disadvantages dynamically (a 'ratchet' effect).
• * This dynamic injustice stems from the distribution of a beneficial good (AI technology) itself.
• * Training data concentration further disadvantages under-represented groups by impacting AI tool quality, not just access.

Conclusions:

• * AI in medicine, under current governance, creates structural injustice, not simple maldistribution.
• * Iris Marion Young's framework is more adequate for understanding this compounding disadvantage.
• * Ethical considerations must address the dynamics of AI diffusion and professional responsibility.