Position: Beyond Sensitive Attributes, ML Fairness Should Quantify Structural Injustice via Social Determinants
A new position paper argues that the field of algorithmic fairness needs to expand its focus beyond sensitive attributes to quantify structural injustice through social determinants. The authors contend that current technical approaches often overlook how contextual variables can create systemic unfairness, treating them as noise rather than signal. They propose a shift towards auditing structural injustice via social determinants before implementing mitigation strategies, highlighting potential new forms of injustice introduced by attribute-centric methods. AI
IMPACT This research could lead to more robust and equitable AI systems by addressing systemic biases beyond individual attributes.