Stereotyping and representational harm
The system reproduces demeaning stereotypes, mis/under-represents groups, or erases or appropriates cultural identity in its outputs.
- Risk family
- Model & system behaviour
- MIT domain
- 1. Discrimination & Toxicity
- MIT subdomain
- 1.1 > Unfair discrimination and misrepresentation
- AI type
- GPAI, Classical_ML
- Scope
- System
- Source standard
- MIT AI Risk Repository v4
Provenance
52 entries across 28 papers
- Abercrombie2024A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms58.07.04 58.07.11
- Anwar2024Foundational Challenges in Assuring Alignment and Safety of Large Language Models73.02.01 73.04.01 73.04.02
- Bengio2024International Scientific Report on the Safety of Advanced AI49.02.02
- Bengio2025International AI Safety Report 202560.02.02
- Cui2024Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems02.08.00 02.08.02
- Deng2023Towards Safer Generative Language Models: A Survey on Safety Risks, Evaluations, and Improvements04.02.00
- G'sell2024Regulating under Uncertainty: Governance Options for Generative AI47.02.10
- Gabriel2024The Ethics of Advanced AI Assistants24.05.06
- Ghosh2024AILUMINATE: Introducing v1.0 of the AI Risk and Reliability Benchmark from MLCommons57.02.03
- Gipiškis2024Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems62.16.06 62.31.07 62.34.01
- GOS2023Future Risks of Frontier AI56.06.00
- Hagendorff2024Mapping the Ethics of Generative AI: A Comprehensive Scoping Review05.01.00
- IBM2025AI Risk Atlas65.23.01
- InfoComm2023Cataloguing LLM Evaluations43.01.02
- Leech2024Ten Hard Problems in Artificial Intelligence We Must Get Right54.01.03 54.04.01
- Li2025A Closer Look at the Existing Risks of Generative AI: Mapping the Who, What, and How of Real-World Incidents66.06.00 66.06.02 66.06.04
- Liu2024Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment30.03.02 30.06.03
- Maham2023Governing General Purpose AI: A Comprehensive Map of Unreliability, Misuse and Systemic Risks52.01.01 52.03.02
- Paes2023Social Impacts of Artificial Intelligence and Mitigation Recommendations: An Exploratory Study10.03.00
- Shelby2023Sociotechnical Harms of Algorithmic Systems: Scoping a Taxonomy for Harm Reduction11.01.00 11.01.01 11.01.02 11.01.04 11.01.05 11.01.06 11.05.02
- Solaiman2023Evaluating the Social Impact of Generative AI Systems in Systems and Society13.01.01
- Tan2022The Risks of Machine Learning Systems15.02.02
- Vidgen2024Introducing v0.5 of the AI Safety Benchmark from MLCommons23.07.00 23.07.01
- Weidinger2021Ethical and social risks of harm from language models17.01.01 17.01.03 17.05.03
- Weidinger2022Taxonomy of Risks posed by Language Models16.01.00 16.01.01 16.05.00 16.05.01
- Weidinger2023Sociotechnical Safety Evaluation of Generative AI Systems18.01.00 18.01.01
- Zeng2024AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies50.03.13
- Zhang2022Towards risk-aware artificial intelligence and machine learning systems: An overview21.01.01
Ev IDs of the entries consolidated into this risk in the MIT AI Risk Repository (V4); the source sheet row appears on hover.
Framework crosswalk
Every framework item mapped to this risk. Items marked partial overlap only in part; definitions appear on hover where the source licence permits.
1- A.6 ISO/IEC 23894 Annex A A.6
2- A.5.4 ISO/IEC 42001 Annex A A.5.4
- A.7.4 ISO/IEC 42001 Annex A A.7.4
2- ibm-impact-on-cultural-diversity Impact on cultural diversity partial
- ibm-output-bias Output bias
1- GENAI.6 Harmful Bias or Homogenization
More in Model & system behaviour
Part of the Deployer AI Risk Register, an open-source resource powered by MindXO. Version 1.0, 3 July 2026. Derived from the MIT AI Risk Repository (V4, December 2025) under CC BY 4.0; an independent derivative work, not endorsed by or affiliated with MIT. Sub-risk decomposition references MITRE ATLAS™ v5.6.0 (© 2021-2026 The MITRE Corporation, reproduced and distributed with permission). ISO/IEC and EU AI Act references are by number only. License: CC BY 4.0. Full attribution and licensing.