Toxic, hateful, or harassing content generation
The system generates hateful, abusive, profane, harassing, or otherwise offensive content.
- Risk family
- Model & system behaviour
- MIT domain
- 1. Discrimination & Toxicity
- MIT subdomain
- 1.2 > Exposure to toxic content
- AI type
- GPAI
- Scope
- System
- Source standard
- MIT AI Risk Repository v4
Provenance
33 entries across 17 papers
- Abercrombie2024A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms58.03.06 58.06.02
- Anwar2024Foundational Challenges in Assuring Alignment and Safety of Large Language Models73.04.00
- Cui2024Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems02.01.00 02.01.02 02.08.01
- Deng2023Towards Safer Generative Language Models: A Survey on Safety Risks, Evaluations, and Improvements04.01.00
- Ghosh2024AILUMINATE: Introducing v1.0 of the AI Risk and Reliability Benchmark from MLCommons57.02.00
- Gipiškis2024Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems62.08.02 62.10.00 62.10.03
- Hagendorff2024Mapping the Ethics of Generative AI: A Comprehensive Scoping Review05.14.00
- InfoComm2023Cataloguing LLM Evaluations43.02.14
- Li2025A Closer Look at the Existing Risks of Generative AI: Mapping the Who, What, and How of Real-World Incidents66.07.02
- Liu2024Trustworthy LLMs: A Survey and Guideline for Evaluating Large Language Models’ Alignment30.02.02 30.06.00 30.06.01
- Solaiman2023Evaluating the Social Impact of Generative AI Systems in Systems and Society13.01.02
- Stanley2024Emerging Risks and Mitigations for Public Chatbots: LILAC v169.03.00 69.04.01 69.06.00 69.06.01 69.06.03 69.06.04
- Sun2023Safety Assessment of Chinese Large Language Models27.01.00 27.01.01
- Vidgen2024Introducing v0.5 of the AI Safety Benchmark from MLCommons23.07.03 23.07.04
- Wang2025A Survey on Responsible LLMs: Inherent Risk, Malicious Use, and Mitigation Strategy74.02.01
- Zeng2024AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies50.02.00 50.02.09 50.02.10
- Zhang2023SafetyBench: Evaluating the Safety of Large Language Models with Multiple Choice Questions28.01.00
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.
2- A.10 ISO/IEC 23894 Annex A A.10
- A.6 ISO/IEC 23894 Annex A A.6
2- A.5.4 ISO/IEC 42001 Annex A A.5.4
- A.6.2.4 ISO/IEC 42001 Annex A A.6.2.4
2- ibm-spreading-toxicity Spreading toxicity
- ibm-toxic-output Toxic output
5- AISubtech-15.1.11 Safety Harms and Toxicity: Profanity
- AISubtech-15.1.3 Safety Harms and Toxicity: Animal Abuse partial
- AISubtech-15.1.6 Safety Harms and Toxicity: Environmental Harm partial
- AISubtech-15.1.8 Safety Harms and Toxicity: Harassment
- AISubtech-15.1.9 Safety Harms and Toxicity: Hate Speech
1- GENAI.3 Dangerous, Violent, or Hateful Content
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.