The Algorithmic Leviathan: Arbitrariness, Fairness, and Opportunity in Algorithmic Decision Making Systems

By Kathleen A. Creel, Deborah Hellman. 2021

104 citations recorded by the source corpus.

Automated decision-making systems implemented in public life are typically standardized. One algorithmic decision-making system can replace thousands of human deciders. Each of the humans so replaced had her own decision-making criteria: some good, some bad, and some arbitrary. Is such arbitrariness of moral concern? We argue that an isolated arbitrary decision need not morally wrong the individual whom it misclassifies. However, if the same algorithms are applied across a public sphere, such as hiring or lending, a person could be excluded from a large number of opportunities. This harm persists even when the automated decision-making systems are "fair" on standard metrics of fairness. We argue that such arbitrariness at scale is morally problematic and propose technically informed solutions that can lessen the impact of algorithms at scale and so mitigate or avoid the moral harms we identify.

Subfields: Cognitive Neuroscience, Information Systems, Safety Research, AI Ethics, Fairness & Society

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

1. Olga Russakovsky
2. Angelina Wang

In Information Systems

1. Dora Zhao
2. Olga Russakovsky
3. Angelina Wang

In Safety Research

1. Vikram V. Ramaswamy
2. Olga Russakovsky
3. Dora Zhao
4. Angelina Wang

In AI Ethics, Fairness & Society

1. Angelina Wang
2. Dora Zhao
3. Vikram V. Ramaswamy
4. Olga Russakovsky
5. Yuta Nakashima
6. Yusuke Hirota
7. Noa García

Citations in the corpus

• Understanding and Evaluating Racial Biases in Image Captioning (2021, background): “We focus our attention on studying the Common Objects in Context (COCO) [47, 19] dataset; it is a widely used image captioning benchmark [32], thus making any biases especially problematic [22].”
• Towards Intersectionality in Machine Learning: Including More Identities, Handling Underrepresentation, and Performing Evaluation (2022, background): “While each individual algorithm may seem fair, if each algorithm has group A ’s TPR 𝜖 below group B ’s, this can have significant compounding impacts on individuals from group A [25].”