# The Unfairness of Fair Machine Learning: Levelling down and strict egalitarianism by default

arXiv: https://arxiv.org/abs/2302.02404 · HTML: https://mlsanity.com/paper/2302.02404/

## Key facts

- Cited by 36 papers in the map (data through 14 Oct 2026).
- Builds on 0 of its 10 references that are in the map.
- First posted on arXiv 5 Feb 2023.

## Built on by (most cited first, 30 of 36)

- [On (assessing) the fairness of risk score models](https://mlsanity.com/paper/2302.08851/) (2023)
- [OxonFair: A Flexible Toolkit for Algorithmic Fairness](https://mlsanity.com/paper/2407.13710/) (2024)
- [The Role of Subgroup Separability in Group-Fair Medical Image Classification](https://mlsanity.com/paper/2307.02791/) (2023)
- [Rethinking Fair Representation Learning for Performance-Sensitive Tasks](https://mlsanity.com/paper/2410.04120/) (2024)
- [The Case for Globalizing Fairness: A Mixed Methods Study on Colonialism, AI, and Health in Africa](https://mlsanity.com/paper/2403.03357/) (2024)
- [On the Fairness ROAD: Robust Optimization for Adversarial Debiasing](https://mlsanity.com/paper/2310.18413/) (2023)
- [Agent Benchmarks Fail Public Sector Requirements](https://mlsanity.com/paper/2601.20617/) (2026)
- [Synthetic Data Generation for Intersectional Fairness by Leveraging Hierarchical Group Structure](https://mlsanity.com/paper/2405.14521/) (2024)
- [Open Challenges on Fairness of Artificial Intelligence in Medical Imaging Applications](https://mlsanity.com/paper/2407.16953/) (2024)
- [Positive-Sum Fairness: Leveraging Demographic Attributes to Achieve Fair AI Outcomes Without Sacrificing Group Gains](https://mlsanity.com/paper/2409.19940/) (2024)
- [Talking About the Assumption in the Room](https://mlsanity.com/paper/2502.13268/) (2025)
- [Critical Appraisal of Fairness Metrics in Clinical Predictive AI](https://mlsanity.com/paper/2506.17035/) (2025)
- [The Homogenization Problem in LLMs: Towards Meaningful Diversity in AI Safety](https://mlsanity.com/paper/2601.06116/) (2026)
- [When mitigating bias is unfair: multiplicity and arbitrariness in algorithmic group fairness](https://mlsanity.com/paper/2302.07185/) (2023)
- [A multidomain relational framework to guide institutional AI research and adoption](https://mlsanity.com/paper/2303.10106/) (2023)
- [Fair Without Leveling Down: A New Intersectional Fairness Definition](https://mlsanity.com/paper/2305.12495/) (2023)
- [Reranking individuals: The effect of fair classification within-groups](https://mlsanity.com/paper/2401.13391/) (2024)
- [Resource-constrained Fairness](https://mlsanity.com/paper/2406.01290/) (2024)
- [Privilege Scores](https://mlsanity.com/paper/2502.01211/) (2025)
- [Federated learning, ethics, and the double black box problem in medical AI](https://mlsanity.com/paper/2504.20656/) (2025)
- [Software Fairness Dilemma: Is Bias Mitigation a Zero-Sum Game?](https://mlsanity.com/paper/2508.03323/) (2025)
- [A Primer on Causal and Statistical Dataset Biases for Fair and Robust Image Analysis](https://mlsanity.com/paper/2509.04295/) (2025)
- [OxEnsemble: Fair Ensembles for Low-Data Classification](https://mlsanity.com/paper/2512.09665/) (2025)
- [Globalizing Fairness Attributes in Machine Learning: A Case Study on Health in Africa](https://mlsanity.com/paper/2304.02190/) (2023)
- [Unfair Utilities and First Steps Towards Improving Them](https://mlsanity.com/paper/2306.00636/) (2023)
- [NLPGuard: A Framework for Mitigating the Use of Protected Attributes by NLP Classifiers](https://mlsanity.com/paper/2407.01697/) (2024)
- [Thorns and Algorithms: Navigating Generative AI Challenges Inspired by Giraffes and Acacias](https://mlsanity.com/paper/2407.11360/) (2024)
- [Fairness Interventions in Classification: A Study on AI Explainability](https://mlsanity.com/paper/2407.14766/) (2024)
- [Analyzing Fairness of Computer Vision and Natural Language Processing Models](https://mlsanity.com/paper/2412.09900/) (2024)
- [It's complicated. The relationship of algorithmic fairness and non-discrimination provisions for high-risk systems in the EU AI Act](https://mlsanity.com/paper/2501.12962/) (2025)

## Abstract (arXiv)

In recent years fairness in machine learning (ML) has emerged as a highly active area of research and development. Most define fairness in simple terms, where fairness means reducing gaps in performance or outcomes between demographic groups while preserving as much of the accuracy of the original system as possible. This oversimplification of equality through fairness measures is troubling. Many current fairness measures suffer from both fairness and performance degradation, or "levelling down," where fairness is achieved by making every group worse off, or by bringing better performing groups down to the level of the worst off. When fairness can only be achieved by making everyone worse off in material or relational terms through injuries of stigma, loss of solidarity, unequal concern, and missed opportunities for substantive equality, something would appear to have gone wrong in translating the vague concept of 'fairness' into practice. This paper examines the causes and prevalence of levelling down across fairML, and explore possible justifications and criticisms based on philosophical and legal theories of equality and distributive justice, as well as equality law jurisprudence. We find that fairML does not currently engage in the type of measurement, reporting, or analysis necessary to justify levelling down in practice. We propose a first step towards substantive equality in fairML: "levelling up" systems by design through enforcement of minimum acceptable harm thresholds, or "minimum rate constraints," as fairness constraints. We likewise propose an alternative harms-based framework to counter the oversimplified egalitarian framing currently dominant in the field and push future discussion more towards substantive equality opportunities and away from strict egalitarianism by default. N.B. Shortened abstract, see paper for full abstract.