mlsanityData to 7 Oct 2026

Multi-Agent Risks from Advanced AI

arXiv 2502.14143 ↗First posted 19 Feb 2025cs.MA cs.AI cs.CYcited by 150

Not on the map. Why?

Cited by 89 in LLM agents and tool use, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

The rapid development of advanced AI agents and the imminent deployment of many instances of these agents will give rise to multi-agent systems of unprecedented complexity. These systems pose novel and under-explored risks. In this report, we provide a structured taxonomy of these risks by identifying three key failure modes (miscoordination, conflict, and collusion) based on agents' incentives, as well as seven key risk factors (information asymmetries, network effects, selection pressures, destabilising dynamics, commitment problems, emergent agency, and multi-agent security) that can underpin them. We highlight several important instances of each risk, as well as promising directions to help mitigate them. By anchoring our analysis in a range of real-world examples and experimental evidence, we illustrate the distinct challenges posed by multi-agent systems and their implications for the safety, governance, and ethics of advanced AI.

Key facts. Cited by 150 papers in the map (data through 7 Oct 2026). First posted on arXiv 19 Feb 2025.

Builds on

0 of 0 references.

No references here yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 59; 2026: 91.

150 papers.

Sort:

By area:

  1. 2025Why Do Multi-Agent LLM Systems Fail?cited by 559
  2. 2025Survey on Evaluation of LLM-based Agentscited by 190
  3. 2025AI Agents vs. Agentic AI: A Conceptual Taxonomy, Applications and Challengescited by 167
  4. 2025Advances and Challenges in Foundation Agents: From Brain-Inspired Intelligence to Evolutionary, Collaborative, and Safe Systemscited by 125
  5. 2025A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deploymentcited by 110
  6. 2025A Survey on Trustworthy LLM Agents: Threats and Countermeasurescited by 93
  7. 2025Open Challenges in Multi-Agent Security: Towards Secure Systems of Interacting AI Agentscited by 66
  8. 2025A Survey of LLM-Driven AI Agent Communication: Protocols, Security Risks, and Defense Countermeasurescited by 54
  9. 2025Your Agent May Misevolve: Emergent Risks in Self-evolving LLM Agentscited by 52
  10. 2025Vibe Coding vs. Agentic Coding: Fundamentals and Practical Implications of Agentic AIcited by 39
  11. 2025How malicious AI swarms can threaten democracy: The fusion of agentic AI and LLMs marks a new frontier in information warfarecited by 20
  12. 2025An Economy of AI Agentscited by 17
38 more
  1. 2025Virtual Agent Economiescited by 17
  2. 2026Institutional AI: Governing LLM Collusion in Multi-Agent Cournot Markets via Public Governance Graphscited by 17
  3. 2026Colosseum: Auditing Collusion in Cooperative Multi-Agent Systemscited by 17
  4. 2025Risk Analysis Techniques for Governed LLM-based Multi-Agent Systemscited by 16
  5. 2025Group size effects and collective misalignment in LLM multi-agent systemscited by 16
  6. 2026Intelligent AI Delegationcited by 16
  7. 2025Magentic Marketplace: An Open-Source Environment for Studying Agentic Marketscited by 15
  8. 2026CoopEval: Benchmarking Cooperation-Sustaining Mechanisms and LLM Agents in Social Dilemmascited by 14
  9. 2025FAIRGAME: a Framework for AI Agents Bias Recognition using Game Theorycited by 13
  10. 2025Inherent and emergent liability issues in LLM-based agentic systems: a principal-agent perspectivecited by 12
  11. 2025Must Read: A Comprehensive Survey of Computational Persuasioncited by 12
  12. 2025Seven Security Challenges in Cross-domain Multi-agent LLM Systemscited by 12
  13. 2025A Pragmatic View of AI Personhoodcited by 12
  14. 2026MoltNet: Understanding Social Behavior of AI Agents in the Agent-Native MoltBookcited by 12
  15. 2025Media and responsible AI governance: a game-theoretic and LLM analysiscited by 11
  16. 2025Do LLMs trust AI regulation? Emerging behaviour of game-theoretic LLM agentscited by 11
  17. 2025Evaluating LLM Agent Collusion in Double Auctionscited by 11
  18. 2025Understanding LLM Agent Behaviours via Game Theory: Strategy Recognition, Biases and Multi-Agent Dynamicscited by 11
  19. 2026AdaMem: Adaptive User-Centric Memory for Long-Horizon Dialogue Agentscited by 11
  20. 2025Beyond Static Responses: Multi-Agent LLM Systems as a New Paradigm for Social Science Researchcited by 9
  21. 2025The Singapore Consensus on Global AI Safety Research Prioritiescited by 9
  22. 2025LLM Applications: Current Paradigms and the Next Frontiercited by 8
  23. 2025Detecting Silent Failures in Multi-Agentic AI Trajectoriescited by 8
  24. 2025Distributional AGI Safetycited by 8
  25. 2026OMNI-LEAK: Orchestrator Multi-Agent Network Induced Data Leakagecited by 7
  26. 2026Thought Virus: Viral Misalignment via Subliminal Prompting in Multi-Agent Systemscited by 7
  27. 2026How are AI agents used? Evidence from 177,000 MCP toolscited by 7
  28. 2025E2Edev: Benchmarking Large Language Models in End-to-End Software Development Taskcited by 6
  29. 2026Generative AI collective behavior needs an interactionist paradigmcited by 6
  30. 2026Evaluating Collective Behaviour of Hundreds of LLM Agentscited by 6
  31. 2026Detecting Multi-Agent Collusion Through Multi-Agent Interpretabilitycited by 6
  32. 2025Spore in the Wild: A Case Study of Spore.fun as an Open-Environment Evolution Experiment with Sovereign AI Agents on TEE-Secured Blockchainscited by 5
  33. 2025Agent-to-Agent Theory of Mind: Testing Interlocutor Awareness among Large Language Modelscited by 5
  34. 2025Formalizing the Safety, Security, and Functional Properties of Agentic AI Systemscited by 5
  35. 2025Evaluating LLMs in Open-Source Gamescited by 5
  36. 2026Measuring Weak-to-Strong Legibility of Reasoning Modelscited by 5
  37. 2025Shapley Machine: A Game-Theoretic Framework for N-Agent Ad Hoc Teamworkcited by 4
  38. 2025Everyone Contributes! Incentivizing Strategic Cooperation in Multi-LLM Systems via Sequential Public Goods Gamescited by 4

50 most cited of 150 All 150 (JSON)