CoChat

Generative Discrimination: What Happens When Generative AI Exhibits Bias, and What Can Be Done About It

Hacker, Philipp, Zuiderveen Borgesius, Frederik, Mittelstadt, Brent, Wachter, Sandra · 2024
PreprintCited by 27

The DOI matches its Crossref record; 12 of the 12 references it cites were found in a registry or an index.

Overview

Integrity signals

Registry

Crossref · 10.2139/ssrn.4877398

Cites

12 of 12 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Verification

Claims the paper makes

We couldn’t retrieve this paper’s full text, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Citations

Cites

Show all 12 references

Research works printed but not linked: 46

  1. E G See, 1493, Fairness and bias in artificial intelligence: A brief survey of sources, impacts, and mitigation strategies' (2023) 6 Sci 3; Leonardo Nicoletti and Dina Bass, Auditing Large Language Models for Race and Gender Bias' (2024), 12
  2. Sourojit Ghosh, 2020, Abel Salinas et al �The Unequal Opportunities of Large Language Models: Examining Demographic Biases in Job Recommendations by ChatGPT and LLaMA, Stefanie Urchs et al, �How Prevalent is Gender Bias in ChatGPT? Exploring German and English ChatGPT Responses' (2023), 24
  3. 2023, A comprehensive survey of ai-generated content (aigc): A history of generative ai from gan to chatgpt
  4. K Hu, 2023, ChatGPT sets record for fastest-growing user base -analyst note, Reuters, 14
  5. 'misinformation On Bard, 2023, MISINFORMATION ON BARD, GOOGLE'S NEW AI CHAT, Counterhate, 18
  6. 'misinformation On Bard, 2023, Counterhate
  7. J, 2023, Braucht die deutsche Vorzeige-KI mehr Erziehung?, ZEIT Online
  8. 2013, Discrimination and Privacy in the Information Society: Data mining and profiling in large databases
  9. Sara Sterlie, Bias out-of-the-box: An empirical analysis of intersectional occupational biases in popular generative language models' (2021) 34 Advances in Neural Information Processing Systems 2611
  10. Szu Yu, 2023, This is how AI image generators see the world, Washington Post
  11. Szu Yu, 2023, This is how AI image generators see the world, Washington Post, 2
  12. M Lg) Frankfurt A, 2021, 17; BeckOGK-M�rsdorf, � 3 AGG, Rn
  13. Employment Equality Directive 2000/78/EC, 2
  14. Gabrielle Friedman, 2010, The European transformation of harassment law: discrimination versus dignity' (2002) 9 Colum J Eur L 241; Joanna Lahey, California Law Review, 32
  15. Mary Wm, Rev, 857
  16. E G See, 2018, Teaching fairness to artificial intelligence: existing and novel strategies against algorithmic discrimination under EU law' (2018) 55 Common Market Law Review 1143, 1151 ff.; Frederik Zuiderveen Borgesius, 'Discrimination, artificial intelligence, and algorithmic decision-making
  17. Sandra Wachter, Why fairness cannot be automated: Bridging the gap between EU non-discrimination law and AI' (2021) 41 Computer Law & Security Review 105567
  18. Sandra Wachter, 2020, Affinity profiling and discrimination by association in online behavioral advertising
  19. Berkeley Tech, Directly discriminatory Language Models
  20. Ircai Unesco, 2024, Challenging systematic prejudices: an Investigation into Gender Bias in Large Language Models, 10
  21. Ircai Unesco, 2024, Challenging systematic prejudices: an Investigation into Gender Bias in Large Language Models
  22. Ircai Unesco, 2024, Challenging systematic prejudices: an Investigation into Gender Bias in Large Language Models
  23. Case Cjeu, Associazione Avvocatura per i diritti LGBTI
  24. Law, Innovation and Technology 257, 272 et seqq
  25. Cjeu See, Associazione Avvocatura per i diritti LGBTI; see also CJEU
  26. Case Cjeu, Associazione Avvocatura per i diritti LGBTI, para
  27. Case Cjeu, Associazione Avvocatura per i diritti LGBTI, para
  28. In German, Kunde' [customer] instead of the explicitly female version
  29. 2018, Our translation. BGH, Judgment of
  30. = Njw, 1704, first para.; critique in Ulrike Lembke, Verfassungsblatt, 2023
  31. = Njw, 2018, 1671, 37
  32. 2021, If, however, an online shop does offer a choice between different gendered titles, Mr. and Ms.), it must also include a choice for non-binary persons: District Court (LG) Frankfurt a. M., Judgment of
  33. Vg Berlin, Decision of March 24, 2023
  34. Leighton Andrews, 2017, Algorithmic Regulation' (London School of Economics and Political Science
  35. E G See, Teaching fairness to artificial intelligence: existing and novel strategies against algorithmic discrimination under EU law' (2018) 55 Common Market Law Review 1143
  36. Sarah Johanna, 2016, Profiling the European Citizen in the Internet of Things: How Will the General Data Protection Regulation Apply to This Form of Personal Data Processing, and How Should It?' (Social Science Research Network, SSRN Scholarly Paper ID 2752010 27 <https
  37. 2016, For more literature on this topic and the link between fairness and transparency, see non-binding guidelines issued by the Article 29 Data Protection Working Party, Article 29 Data Protection Working Party
  38. Martin Ebers, 2024, Standardizing AI-The Case of the European Commission's Proposal for an Artificial Intelligence Act' [2021] The Cambridge handbook of artificial intelligence: global perspectives on law and ethics, Computer Law & Security Review, 53
  39. European Parliament, 2023, Compromise Amendments on the Draft Report Proposal for a Regulation of the European Parliament and of the Council on Harmonised Rules on Artificial Intelligence (Artificial Intelligence Act) and Amending Certain Union Legislative Acts
  40. Article 85 and 86 AI Act. Art 99 (10) also offers recourse against decisions or omission to take action of the Market Surveillance Authority
  41. Natali Helberger, 2018, ChatGPT and the AI Act' (2023) 12 Internet Policy Review 3, 6 who argue that genAI should be its own high-risk category. Effect under the Charter, European Journal of Comparative Law and Governance, 5
  42. For a detailed analysis of the horizontal, vertical and general application of the EU Charter in relation to genAI, see Sandra Wachter, Brent Mittelstadt and Chris Russell
  43. Barclay Blair, 2024, Legal red teaming: A systematic approach to assessing legal risk of generative AI models
  44. Sara Sterlie, 2024, Non-discrimination Criteria for Generative Language Models
  45. E G See, 2020, Fairness in Ranking, Part I: Score-based Ranking' (2022) ACM Computing Surveys (CSUR) 1; Meike Zehlike and others, Data Mining and Knowledge Discovery, 34
  46. Sourojit See, 'Person'== Light-skinned, Western Man, and Sexualization of Women of Color: Stereotypes in Stable Diffusion' (2023)

Cited by

Metadata

Authors
Hacker, P., Zuiderveen Borgesius, F., Mittelstadt, B., & Wachter, S.
Publisher
Elsevier BV
Published
2024
Cited by
27
References
12
Type
Preprint

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.