# AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies

arXiv: https://arxiv.org/abs/2406.17864 · HTML: https://mlsanity.com/paper/2406.17864/

## Key facts

- Cited by 37 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 61 references that are in the map.
- First posted on arXiv 25 Jun 2024.

## Builds on (most cited first, 13 of 13)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Universal and Transferable Adversarial Attacks on Aligned Language Models](https://mlsanity.com/paper/2307.15043/) (2023)
- [HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal](https://mlsanity.com/paper/2402.04249/) (2024)
- [RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models](https://mlsanity.com/paper/2009.11462/) (2020)
- [Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!](https://mlsanity.com/paper/2310.03693/) (2023)
- [Ethical and social risks of harm from Language Models](https://mlsanity.com/paper/2112.04359/) (2021)
- [JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Models](https://mlsanity.com/paper/2404.01318/) (2024)
- [DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models](https://mlsanity.com/paper/2306.11698/) (2023)
- [SALAD-Bench: A Hierarchical and Comprehensive Safety Benchmark for Large Language Models](https://mlsanity.com/paper/2402.05044/) (2024)
- [Stability](https://mlsanity.com/paper/1310.0150/) (2013)
- [Behavioral Use Licensing for Responsible AI](https://mlsanity.com/paper/2011.03116/) (2020)

## Built on by (most cited first, 30 of 37)

- [OpenAgentSafety: A Comprehensive Framework for Evaluating Real-World AI Agent Safety](https://mlsanity.com/paper/2507.06134/) (2025)
- [AIR-Bench 2024: A Safety Benchmark Based on Risk Categories from Regulations and Policies](https://mlsanity.com/paper/2407.17436/) (2024)
- [Large Language Model Safety: A Holistic Survey](https://mlsanity.com/paper/2412.17686/) (2024)
- [AutoRedTeamer: Autonomous Red Teaming with Lifelong Attack Integration](https://mlsanity.com/paper/2503.15754/) (2025)
- [Machine Unlearning Doesn't Do What You Think: Lessons for Generative AI Policy and Research](https://mlsanity.com/paper/2412.06966/) (2024)
- [Steering the CensorShip: Uncovering Representation Vectors for LLM "Thought" Control](https://mlsanity.com/paper/2504.17130/) (2025)
- [Acceptable Use Policies for Foundation Models](https://mlsanity.com/paper/2409.09041/) (2024)
- [Survey on AI-Generated Media Detection: From Non-MLLM to MLLM](https://mlsanity.com/paper/2502.05240/) (2025)
- [SafetyAnalyst: Interpretable, Transparent, and Steerable Safety Moderation for AI Behavior](https://mlsanity.com/paper/2410.16665/) (2024)
- [MinorBench: A hand-built benchmark for content-based risks for children](https://mlsanity.com/paper/2503.10242/) (2025)
- [Evolve the Method, Not the Prompts: Evolutionary Synthesis of Jailbreak Attacks on LLMs](https://mlsanity.com/paper/2511.12710/) (2025)
- [Taxonomizing Representational Harms using Speech Act Theory](https://mlsanity.com/paper/2504.00928/) (2025)
- [Bench-2-CoP: Can We Trust Benchmarking for EU AI Compliance?](https://mlsanity.com/paper/2508.05464/) (2025)
- [Refusal Steering: Fine-grained Control over LLM Refusal Behaviour for Sensitive Topics](https://mlsanity.com/paper/2512.16602/) (2025)
- [Class-RAG: Real-Time Content Moderation with Retrieval Augmented Generation](https://mlsanity.com/paper/2410.14881/) (2024)
- [BeSafe-Bench: Unveiling Behavioral Safety Risks of Situated Agents in Functional Environments](https://mlsanity.com/paper/2603.25747/) (2026)
- [Data-Centric AI Governance: Addressing the Limitations of Model-Focused Policies](https://mlsanity.com/paper/2409.17216/) (2024)
- [Risk Sources and Risk Management Measures in Support of Standards for General-Purpose AI Systems](https://mlsanity.com/paper/2410.23472/) (2024)
- [SAIF: A Comprehensive Framework for Evaluating the Risks of Generative AI in the Public Sector](https://mlsanity.com/paper/2501.08814/) (2025)
- [Breaking Down Bias: On The Limits of Generalizable Pruning Strategies](https://mlsanity.com/paper/2502.07771/) (2025)
- [Envisioning Stakeholder-Action Pairs to Mitigate Negative Impacts of AI: A Participatory Approach to Inform Policy Making](https://mlsanity.com/paper/2502.14869/) (2025)
- [A Systematic Review of Open Datasets Used in Text-to-Image (T2I) Gen AI Model Safety](https://mlsanity.com/paper/2503.00020/) (2025)
- [Interpreting LLM-as-a-Judge Policies via Verifiable Global Explanations](https://mlsanity.com/paper/2510.08120/) (2025)
- [Agent-Supported Foresight for AI Systemic Risks: AI Agents for Breadth, Experts for Judgment](https://mlsanity.com/paper/2602.08565/) (2026)
- [Brainrot: Deskilling and Addiction are Overlooked AI Risks](https://mlsanity.com/paper/2605.03512/) (2026)
- [Dimensional Characterization and Pathway Modeling for Catastrophic AI Risks](https://mlsanity.com/paper/2508.06411/) (2025)
- [Responsible AI Technical Report](https://mlsanity.com/paper/2509.20057/) (2025)
- [Standardization Trends on Safety and Trustworthiness Technology for Advanced AI](https://mlsanity.com/paper/2410.22151/) (2024)
- [ACCESS DENIED INC: The First Benchmark Environment for Sensitivity Awareness](https://mlsanity.com/paper/2506.00964/) (2025)
- [Guard Vector: Beyond English LLM Guardrails with Task-Vector Composition and Streaming-Aware Prefix SFT](https://mlsanity.com/paper/2509.23381/) (2025)

## Abstract (arXiv)

We present a comprehensive AI risk taxonomy derived from eight government policies from the European Union, United States, and China and 16 company policies worldwide, making a significant step towards establishing a unified language for generative AI safety evaluation. We identify 314 unique risk categories organized into a four-tiered taxonomy. At the highest level, this taxonomy encompasses System & Operational Risks, Content Safety Risks, Societal Risks, and Legal & Rights Risks. The taxonomy establishes connections between various descriptions and approaches to risk, highlighting the overlaps and discrepancies between public and private sector conceptions of risk. By providing this unified framework, we aim to advance AI safety through information sharing across sectors and the promotion of best practices in risk mitigation for generative AI models and systems.