mlsanityData to 14 Oct 2026

Characterizing the risk of fairwashing

arXiv 2106.07504 ↗First posted 14 Jun 2021cs.LG cs.CYcited by 21

Not on the map. Why?

Abstract (arXiv)

Fairwashing refers to the risk that an unfair black-box model can be explained by a fairer model through post-hoc explanation manipulation. In this paper, we investigate the capability of fairwashing attacks by analyzing their fidelity-unfairness trade-offs. In particular, we show that fairwashed explanation models can generalize beyond the suing group (i.e., data points that are being explained), meaning that a fairwashed explainer can be used to rationalize subsequent unfair decisions of a black-box model. We also demonstrate that fairwashing attacks can transfer across black-box models, meaning that other black-box models can perform fairwashing without explicitly using their predictions. This generalization and transferability of fairwashing attacks imply that their detection will be difficult in practice. Finally, we propose an approach to quantify the risk of fairwashing, which is based on the computation of the range of the unfairness of high-fidelity explainers.

Key facts. Cited by 21 papers in the map (data through 14 Oct 2026). Builds on 35 of its 54 references that are in the map. First posted on arXiv 14 Jun 2021.

Builds on

35 of 54 references.

References · 35

  1. 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
  2. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,166
  3. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,070
  4. 2016Equality of Opportunity in Supervised Learningcited by 2,007
  5. 2019Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AIcited by 1,552
  6. 2018Sanity Checks for Saliency Mapscited by 1,156
  7. 2018A Survey Of Methods For Explaining Black Box Modelscited by 973
  8. 2016The Mythos of Model Interpretabilitycited by 721
  9. 2014Certifying and removing disparate impactcited by 715
  10. 2016Fair prediction with disparate impact: A study of bias in recidivism prediction instrumentscited by 672
  11. 2018A Reductions Approach to Fair Classificationcited by 587
  12. 2016Inherent Trade-Offs in the Fair Determination of Risk Scorescited by 586
23 more
  1. 2020Underspecification Presents Challenges for Credibility in Modern Machine Learningcited by 496
  2. 2019Fooling LIME and SHAP: Adversarial Attacks on Post hoc Explanation Methodscited by 443
  3. 2017On Fairness and Calibrationcited by 406
  4. 2017Algorithmic decision making and the cost of fairnesscited by 403
  5. 2016Fairness Beyond Disparate Treatment & Disparate Impact: Learning Classification without Disparate Mistreatmentcited by 369
  6. 2021Interpretable Machine Learning: Fundamental Principles and 10 Grand Challengescited by 367
  7. 2019Explanations can be manipulated and geometry is to blamecited by 216
  8. 2019Predictive Multiplicity in Classificationcited by 166
  9. 2020Debugging Tests for Model Explanationscited by 147
  10. 2019Fooling Neural Network Interpretations via Adversarial Model Manipulationcited by 112
  11. 2019"How do I fool you?": Manipulating User Trust via Misleading Black Box Explanationscited by 111
  12. 2019The Dangers of Post-hoc Interpretability: Unjustified Counterfactual Explanationscited by 97
  13. 2018Model Reconstruction from Model Explanationscited by 86
  14. 2019Fairwashing: the risk of rationalizationcited by 81
  15. 2020Fairwashing Explanations with Off-Manifold Detergentcited by 72
  16. 2021Characterizing Fairness Over the Set of Good Models Under Selective Labelscited by 65
  17. 2017Learning Certifiably Optimal Rule Lists for Categorical Datacited by 50
  18. 2015On the relation between accuracy and fairness in binary classificationcited by 48
  19. 2020Model extraction from counterfactual explanationscited by 47
  20. 2020Ensuring Fairness Beyond the Training Datacited by 35
  21. 2019Stable and Fair Classificationcited by 25
  22. 2019Faking Fairness via Stealthily Biased Samplingcited by 14
  23. 2019Learning Fair Rule Listscited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 1; 2022: 3; 2023: 10; 2024: 0; 2025: 4; 2026: 3.

21 papers.

Sort:

  1. 2023Connecting the Dots in Trustworthy Artificial Intelligence: From AI Principles, Ethics, and Key Requirements to Responsible AI Systems and Regulationcited by 61
  2. 2023Adversarial attacks and defenses in explainable artificial intelligence: A surveycited by 48
  3. 2022The Road to Explainability is Paved with Bias: Measuring the Fairness of Explanationscited by 41
  4. 2023A Path to Simpler Models Starts With Noisecited by 33
  5. 2023A Critical Survey on Fairness Benefits of Explainable AIcited by 16
  6. 2022Computing the Collection of Good Models for Rule Listscited by 12
  7. 2021When and How to Fool Explainable Models (and Humans) with Adversarial Examplescited by 10
  8. 2023Why is plausibility surprisingly problematic as an XAI criterion?cited by 9
  9. 2023Adversarial Attacks on the Interpretation of Neuron Activation Maximizationcited by 7
  10. 2023Manipulation Risks in Explainable AI: The Implications of the Disagreement Problemcited by 7
  11. 2023Probabilistic Dataset Reconstruction from Interpretable Modelscited by 7
  12. 2025Robust ML Auditing using Prior Knowledgecited by 6
9 more
  1. 2023Learning Hybrid Interpretable Models: Theory, Taxonomy, and Methodscited by 5
  2. 2023SoK: Taming the Triangle -- On the Interplays between Fairness, Interpretability and Privacy in Machine Learningcited by 4
  3. 2022Tensions Between the Proxies of Human Values in AIcited by 0
  4. 2025A Critical Review of Predominant Bias in Neural Networkscited by 0
  5. 2025Fair Play for Individuals, Foul Play for Groups? Auditing Anonymization's Impact on ML Fairnesscited by 0
  6. 2025Breaking the Illusion of Security via Interpretation: Interpretable Vision Transformer Systems under Attackcited by 0
  7. 2026On the Definition and Detection of Cherry-Picking in Counterfactual Explanationscited by 0
  8. 2026Mixtures of Transparent Local Modelscited by 0
  9. 2026Model Stealing Through the Lens of Model Multiplicitycited by 0