mlsanityData to 14 Oct 2026

Dimensional Characterization and Pathway Modeling for Catastrophic AI Risks

arXiv 2508.06411 ↗First posted 8 Aug 2025cs.CY cs.AI cs.LGcited by 1

Not on the map. Why?

Abstract (arXiv)

Although discourse around the risks of Artificial Intelligence (AI) has grown, it often lacks a comprehensive, multidimensional framework, and concrete causal pathways mapping hazard to harm. This paper aims to bridge this gap by examining six commonly discussed AI catastrophic risks: CBRN, cyber offense, sudden loss of control, gradual loss of control, environmental risk, and geopolitical risk. First, we characterize these risks across seven key dimensions, namely intent, competency, entity, polarity, linearity, reach, and order. Next, we conduct risk pathway modeling by mapping step-by-step progressions from the initial hazard to the resulting harms. The dimensional approach supports systematic risk identification and generalizable mitigation strategies, while risk pathway models help identify scenario-specific interventions. Together, these methods offer a more structured and actionable foundation for managing catastrophic AI risks across the value chain.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 63 of its 140 references that are in the map. First posted on arXiv 8 Aug 2025.

Builds on

63 of 140 references.

References · 63

  1. 2023An Overview of Catastrophic AI Riskscited by 222
  2. 2021Trustworthy AI: From Principles to Practicescited by 207
  3. 2023Model evaluation for extreme riskscited by 177
  4. 2018The Malicious Use of Artificial Intelligence: Forecasting, Prevention, and Mitigationcited by 165
  5. 2025Multi-Agent Risks from Advanced AIcited by 163
  6. 2023Harms from Increasingly Agentic Algorithmic Systemscited by 135
  7. 2024Teams of LLM Agents can Exploit Zero-Day Vulnerabilitiescited by 95
  8. 2024Visibility into AI Agentscited by 79
  9. 2025Superintelligent Agents Pose Catastrophic Risks: Can Scientist AI Offer a Safer Path?cited by 75
  10. 2025Artificial Intelligence Index Report 2025cited by 64
  11. 2024Open Problems in Technical AI Governancecited by 49
  12. 2025Gradual Disempowerment: Systemic Existential Risks from Incremental AI Developmentcited by 46
38 more
  1. 2023Artificial intelligence and biological misuse: Differentiating risks of language models and biological design toolscited by 45
  2. 2025An Approach to Technical AGI Safety and Securitycited by 45
  3. 2025Infrastructure for AI Agentscited by 41
  4. 2024AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policiescited by 37
  5. 2024Towards evaluations-based safety cases for AI schemingcited by 33
  6. 2023Ecosystem Graphs: The Social Footprint of Foundation Modelscited by 32
  7. 2023TASRA: a Taxonomy and Analysis of Societal-Scale Risks from AIcited by 26
  8. 2024Two Types of AI Existential Risk: Decisive and Accumulativecited by 22
  9. 2023Risk assessment at AGI companies: A review of popular risk assessment techniques from other safety-critical industriescited by 19
  10. 2024A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harmscited by 19
  11. 2025Reality Check: A New Evaluation Ecosystem Is Necessary to Understand AI's Real World Effectscited by 19
  12. 2023Towards more Practical Threat Models in Artificial Intelligence Securitycited by 17
  13. 2022Current and Near-Term AI as a Potential Existential Risk Factorcited by 14
  14. 2024A Taxonomy of Systemic Risks from General-Purpose AIcited by 12
  15. 2024Societal Adaptation to Advanced AIcited by 10
  16. 2016Artificial Intelligence Safety and Cybersecurity: a Timeline of AI Failurescited by 9
  17. 2022A taxonomic system for failure cause analysis of open source AI incidentscited by 9
  18. 2024The Reality of AI and Bioriskcited by 8
  19. 2025A Frontier AI Risk Management Framework: Bridging the Gap Between Current AI Practices and Established Risk Managementcited by 8
  20. 2025AI Governance to Avoid Extinction: The Strategic Landscape and Actionable Research Questionscited by 6
  21. 2025A Unified Framework for Human AI Collaboration in Security Operations Centers with Trusted Autonomycited by 6
  22. 2025Systematic Hazard Analysis for Frontier AI using STPAcited by 6
  23. 2022The Risks of Machine Learning Systemscited by 5
  24. 2025Lessons from complexity theory for AI governancecited by 5
  25. 2025From Incidents to Insights: Patterns of Responsibility following AI Harmscited by 5
  26. 2025Contemporary AI foundation models increase biological weapons riskcited by 5
  27. 2015Taxonomy of Pathways to Dangerous AIcited by 4
  28. 2016A Model of Pathways to Artificial Superintelligence Catastrophe for Risk and Decision Analysiscited by 4
  29. 2024Mapping Technical Safety Research at AI Companies: A literature review and incentives analysiscited by 4
  30. 2024AI, Climate, and Regulation: From Data Centers to the AI Actcited by 4
  31. 2025Examining Popular Arguments Against AI Existential Risk: A Philosophical Analysiscited by 4
  32. 2025Adapting Probabilistic Risk Assessment for AIcited by 4
  33. 2025An Example Safety Case for Safeguards Against Misusecited by 4
  34. 2025Measuring What Matters: A Framework for Evaluating Safety Risks in Real-World LLM Applicationscited by 4
  35. 2023The Ethics of AI Value Chainscited by 3
  36. 2025Towards Adaptive Categories: Dimensional Governance for Agentic AIcited by 3
  37. 2025Mitigating Cyber Risk in the Age of Open-Weight LLMs: Policy Gaps and Technical Realitiescited by 3
  38. 2025Distinguishing Task-Specific and General-Purpose AI in Regulationcited by 3

50 most cited of 63 All 63 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Reframing AI Loss of Control: What Control Is, How to Have It, How to Lose Itcited by 0