mlsanityData to 7 Oct 2026

The Fallacy of AI Functionality

arXiv 2206.09511 ↗First posted 20 Jun 2022cs.LGcited by 88

Not on the map. Why?

Abstract (arXiv)

Deployed AI systems often do not work. They can be constructed haphazardly, deployed indiscriminately, and promoted deceptively. However, despite this reality, scholars, the press, and policymakers pay too little attention to functionality. This leads to technical and policy solutions focused on "ethical" or value-aligned deployments, often skipping over the prior question of whether a given system functions, or provides any benefits at all. To describe the harms of various types of functionality failures, we analyze a set of case studies to create a taxonomy of known AI functionality issues. We then point to policy and organizational responses that are often overlooked and become more readily available once functionality is drawn into focus. We argue that functionality is a meaningful AI policy challenge, operating as a necessary first step towards protecting affected communities from algorithmic harm.

Key facts. Cited by 88 papers in the map (data through 7 Oct 2026). Builds on 38 of its 211 references that are in the map. First posted on arXiv 20 Jun 2022.

Builds on

38 of 211 references.

References · 38

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  2. 2021Ethical and social risks of harm from Language Modelscited by 957
  3. 2014Certifying and removing disparate impactcited by 715
  4. 2019Release Strategies and the Social Impacts of Language Modelscited by 505
  5. 2020Measuring Robustness to Natural Distribution Shifts in Image Classificationcited by 452
  6. 2020Closing the AI Accountability Gap: Defining an End-to-End Framework for Internal Algorithmic Auditingcited by 380
  7. 2021AI and the Everything in the Whole Wide World Benchmarkcited by 340
  8. 2020Improving Reproducibility in Machine Learning Research (A Report from the NeurIPS 2019 Reproducibility Program)cited by 318
  9. 2019What BERT is not: Lessons from a new suite of psycholinguistic diagnostics for language modelscited by 317
  10. 2019Measurement and Fairnesscited by 299
  11. 2019Hidden Stratification Causes Clinically Meaningful Failures in Machine Learning for Medical Imagingcited by 215
  12. 2019Explainable Machine Learning in Deploymentcited by 210
26 more
  1. 2018A comparative study of fairness-enhancing interventions in machine learningcited by 208
  2. 2019Mitigating Bias in Algorithmic Hiring: Evaluating Claims and Practicescited by 170
  3. 2019Combating Adversarial Misspellings with Robust Word Recognitioncited by 162
  4. 2018The Malicious Use of Artificial Intelligence: Forecasting, Prevention, and Mitigationcited by 159
  5. 2020Common pitfalls and recommendations for using machine learning to detect and prognosticate for COVID-19 using chest radiographs and CT scanscited by 158
  6. 2019The Hidden Assumptions Behind Counterfactual Explanations and Principal Reasonscited by 115
  7. 2019Artificial Intelligence: the global landscape of ethics guidelinescited by 107
  8. 2017Runaway Feedback Loops in Predictive Policingcited by 97
  9. 2018Has Machine Translation Achieved Human Parity? A Case for Document-level Evaluationcited by 88
  10. 2021Evaluating the Robustness of Neural Language Models to Input Perturbationscited by 87
  11. 2019Problem Formulation and Fairnesscited by 85
  12. 2021Demystifying the Draft EU Artificial Intelligence Actcited by 82
  13. 2016A Confidence-Based Approach for Balancing Fairness and Accuracycited by 73
  14. 2020The De-democratization of AI: Deep Learning and the Compute Divide in Artificial Intelligence Researchcited by 67
  15. 2018Attaining the Unattainable? Reassessing Claims of Human Parity in Neural Machine Translationcited by 58
  16. 2020Trust in Data Science: Collaboration, Translation, and Accountability in Corporate Data Science Projectscited by 24
  17. 2021Dodging Attack Using Carefully Crafted Natural Makeupcited by 17
  18. 2021A Silicon Valley Love Triangle: Hiring Algorithms, Pseudo-Science, and the Quest for Auditabilitycited by 16
  19. 2019Defining AI in Policy versus Practicecited by 15
  20. 2021The effect of differential victim crime reporting on predictive policing systemscited by 15
  21. 2019ABOUT ML: Annotation and Benchmarking on Understanding and Transparency of Machine Learning Lifecyclescited by 13
  22. 2019Hard Choices in Artificial Intelligence: Addressing Normative Uncertainty through Sociotechnical Commitmentscited by 8
  23. 2021Measurement as governance in and for responsible AIcited by 8
  24. 2020Beyond Near- and Long-Term: Towards a Clearer Account of Research Priorities in AI Ethics and Societycited by 7
  25. 2019MLSys: The New Frontier of Machine Learning Systemscited by 4
  26. 2021Engagement Outweighs Exposure to Partisan and Unreliable News within Google Searchcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 6; 2023: 20; 2024: 25; 2025: 24; 2026: 13.

88 papers.

Sort:

By area:

  1. 2022BLOOM: A 176B-Parameter Open-Access Multilingual Language Modelcited by 1,968
  2. 2024Lessons from the Trenches on Reproducible Evaluation of Language Modelscited by 172
  3. 2024Hallucination-Free? Assessing the Reliability of Leading AI Legal Research Toolscited by 170
  4. 2023Model evaluation for extreme riskscited by 167
  5. 2023Auditing large language models: a three-layered approachcited by 135
  6. 2023Harms from Increasingly Agentic Algorithmic Systemscited by 134
  7. 2023Evaluating and Mitigating Discrimination in Language Model Decisionscited by 117
  8. 2025ELEPHANT: Measuring and understanding social sycophancy in LLMscited by 115
  9. 2024Visibility into AI Agentscited by 77
  10. 2025Authenticated Delegation and Authorized AI Agentscited by 63
  11. 2025Toward an Evaluation Science for Generative AI Systemscited by 42
  12. 2022A Validity Perspective on Evaluating the Justified Use of Data-driven Decision-making Algorithmscited by 30
38 more
  1. 2024Towards interactive evaluations for interaction harms in human-AI systemscited by 27
  2. 2022From plane crashes to algorithmic harm: applicability of safety engineering frameworks for responsible MLcited by 26
  3. 2025Lessons From Red Teaming 100 Generative AI Productscited by 22
  4. 2024LLM-Driven Robots Risk Enacting Discrimination, Violence, and Unlawful Actionscited by 21
  5. 2024A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harmscited by 19
  6. 2024Not My Voice! A Taxonomy of Ethical and Safety Harms of Speech Generatorscited by 18
  7. 2024IDs for AI Systemscited by 18
  8. 2025Reality Check: A New Evaluation Ecosystem Is Necessary to Understand AI's Real World Effectscited by 18
  9. 2024BendVLM: Test-Time Debiasing of Vision-Language Embeddingscited by 17
  10. 2024The "Colonial Impulse" of Natural Language Processing: An Audit of Bengali Sentiment Analysis Tools and Their Identity-based Biasescited by 16
  11. 2022Aligning Offline Metrics and Human Judgments of Value for Code Generation Modelscited by 14
  12. 2023REFORMS: Reporting Standards for Machine Learning Based Sciencecited by 13
  13. 2023"One-Size-Fits-All"? Examining Expectations around What Constitute "Fair" or "Good" NLG System Behaviorscited by 11
  14. 2023Counterfactual Prediction Under Outcome Measurement Errorcited by 10
  15. 2024What is it for a Machine Learning Model to Have a Capability?cited by 10
  16. 2025AI Mismatches: Identifying Potential Algorithmic Harms Before AI Developmentcited by 10
  17. 2023Beneficent Intelligence: A Capability Approach to Modeling Benefit, Assistance, and Associated Moral Failures through AI Systemscited by 9
  18. 2024The Necessity of AI Audit Standards Boardscited by 9
  19. 2024Auditing of AI: Legal, Ethical and Technical Approachescited by 9
  20. 2023The AI Incident Database as an Educational Tool to Raise Awareness of AI Harms: A Classroom Exploration of Efficacy, Limitations, & Future Improvementscited by 8
  21. 2023Toward Operationalizing Pipeline-aware ML Fairness: A Research Agenda for Developing Practical Guidelines and Toolscited by 7
  22. 2023Responsible AI Research Needs Impact Statements Toocited by 7
  23. 2025In-House Evaluation Is Not Enough: Towards Robust Third-Party Flaw Disclosure for General-Purpose AIcited by 7
  24. 2025Rigor in AI: Doing Rigorous AI Work Requires a Broader, Responsible AI-Informed Conception of Rigorcited by 7
  25. 2023Can Workers Meaningfully Consent to Workplace Wellbeing Technologies?cited by 6
  26. 2024Studying Up Public Sector AI: How Networks of Power Relations Shape Agency Decisions Around AI Design and Usecited by 6
  27. 2023Fairness and Sequential Decision Making: Limits, Lessons, and Opportunitiescited by 5
  28. 2023An International Consortium for Evaluations of Societal-Scale Risks from Advanced AIcited by 5
  29. 2023Diffusion Models for Generative Artificial Intelligence: An Introduction for Applied Mathematicianscited by 5
  30. 2024Data Debiasing with Datamodels (D3M): Improving Subgroup Robustness via Data Selectioncited by 5
  31. 2025Real-World Gaps in AI Governance Researchcited by 5
  32. 2023Data AUDIT: Identifying Attribute Utility- and Detectability-Induced Bias in Task Modelscited by 4
  33. 2022Artificial intelligence in government: Concepts, standards, and a unified frameworkcited by 3
  34. 2024Information That Matters: Exploring Information Needs of People Affected by Algorithmic Decisionscited by 3
  35. 2024Unraveling overoptimism and publication bias in ML-driven sciencecited by 3
  36. 2024Legacy Procurement Practices Shape How U.S. Cities Govern AI: Understanding Government Employees' Practices, Challenges, and Needscited by 3
  37. 2025Distinguishing Task-Specific and General-Purpose AI in Regulationcited by 3
  38. 2025A Different Approach to AI Safety: Proceedings from the Columbia Convening on Openness in Artificial Intelligence and AI Safetycited by 3

50 most cited of 88 All 88 (JSON)