mlsanityData to 14 Oct 2026

LLM Agents Make Collective Belief Dynamics Programmable: Challenges and Research Directions

arXiv 2605.19915 ↗First posted 19 May 2026cs.MA cs.SIcited by 0

On the map: LLM agents and tool use → Social simulation and multi-agent debate

Abstract (arXiv)

Classical models of opinion dynamics assume human participants with bounded rationality and limited coordination. The rise of LLM-based agents introduces a qualitative shift: agents can now participate in online discussions at scale, maintain consistent persuasion strategies, and coordinate systematically. This paper argues that LLM agents make collective belief dynamics programmable, enabling deliberate steering of population-level beliefs. We term this emerging problem programmable collective belief control. Through controlled multi-agent simulations, we provide proof-of-concept evidence that coordinated AI agents can induce measurable belief shifts that stabilize within a few interaction rounds. We identify four structural properties (indistinguishability, persistence, contextuality, and configurability) that make detection and defense fundamentally difficult. Based on these findings, we outline a research agenda spanning theoretical foundations for adversarial belief dynamics, operational methods for system-level detection and intervention, and simulation infrastructure for scalable experimentation. Our goal is not to present a complete solution, but to articulate why this problem demands urgent attention and to provide a conceptual foundation for future work.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use. Builds on 48 of its 67 references that are in the map. First posted on arXiv 19 May 2026.

Builds on

48 of 67 references.

Closest ancestors · 4

Same direction, newest first

  1. 2024From Individual to Society: A Survey on Social Simulation Driven by Large Language Model-based Agents4 Dec
  2. 2023Simulating Opinion Dynamics with Networks of LLM-based Agents16 Nov
  3. 2023Epidemic Modeling with Generative Agents11 Jul
  4. 2023Generative Agents: Interactive Simulacra of Human Behavior7 Apr

Other references · 44

  1. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  2. 2022Fast Inference from Transformers via Speculative Decodingcited by 1,653
  3. 2023The Rise and Potential of Large Language Model Based Agents: A Surveycited by 1,439
  4. 2024The AI Scientist: Towards Fully Automated Open-Ended Scientific Discoverycited by 1,071
  5. 2022Red Teaming Language Models with Language Modelscited by 1,026
  6. 2022Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learnedcited by 729
  7. 2023DetectGPT: Zero-Shot Machine-Generated Text Detection using Probability Curvaturecited by 726
  8. 2019Graph Neural Networks for Social Recommendationcited by 714
  9. 2023Paraphrasing evades detectors of AI-generated text, but retrieval is an effective defensecited by 429
  10. 2019AutoML: A Survey of the State-of-the-Artcited by 390
  11. 2023Can AI-Generated Text be Reliably Detected?cited by 386
  12. 2023MLAgentBench: Evaluating Language Agents on Machine Learning Experimentationcited by 366
32 more
  1. 2023Are aligned neural networks adversarially aligned?cited by 315
  2. 2021All That's 'Human' Is Not Gold: Evaluating Human Evaluation of Generated Textcited by 287
  3. 2020Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claimscited by 165
  4. 2024OASIS: Open Agent Social Interaction Simulations with One Million Agentscited by 132
  5. 2023M4: Multi-generator, Multi-domain, and Multi-lingual Black-Box Machine-Generated Text Detectioncited by 121
  6. 2021Parallel Bayesian Optimization of Multiple Noisy Objectives with Expected Hypervolume Improvementcited by 118
  7. 2022Human heuristics for AI-generated language are flawedcited by 105
  8. 2023Combating Misinformation in the Age of LLMs: Opportunities and Challengescited by 95
  9. 2022Diffusion Causal Models for Counterfactual Estimationcited by 87
  10. 2022Evaluating Explainability for Graph Neural Networkscited by 87
  11. 2019Tuning Hyperparameters without Grad Students: Scalable and Robust Bayesian Optimisation with Dragonflycited by 67
  12. 2017Online Human-Bot Interactions: Detection, Estimation, and Characterizationcited by 64
  13. 2023The Gradient of Generative AI Release: Methods and Considerationscited by 63
  14. 2022Adversarial Training for High-Stakes Reliabilitycited by 60
  15. 2019Scalable and Generalizable Social Bot Detection through Data Selectioncited by 59
  16. 2023The Challenge of Using LLMs to Simulate Human Behavior: A Causal Inference Perspectivecited by 59
  17. 2020Uncovering Coordinated Networks on Social Media: Methods and Case Studiescited by 49
  18. 2020Identifying Coordinated Accounts on Social Media through Hidden Influence and Group Behaviourscited by 47
  19. 2023A Survey on Explainability of Graph Neural Networkscited by 29
  20. 2025LLM Strategic Reasoning: Agentic Study through Behavioral Game Theorycited by 17
  21. 2025Leveraging Explainable AI for LLM Text Attribution: Differentiating Human-Written and Multiple LLMs-Generated Textcited by 11
  22. 2023MetaAdapt: Domain Adaptive Few-Shot Misinformation Detection via Meta Learningcited by 8
  23. 2022Opinion dynamics in social networks: From models to datacited by 6
  24. 2017Opinion Polarization by Learning from Social Feedbackcited by 3
  25. 2022Investigating User Radicalization: A Novel Dataset for Identifying Fine-Grained Temporal Shifts in Opinioncited by 3
  26. 2024Evolutionary Multi-agent Reinforcement Learning in Group Social Dilemmascited by 3
  27. 2025Multi-Agent Reinforcement Learning Simulation for Environmental Policy Synthesiscited by 3
  28. 2021Cognitive cascades: How to model (and potentially counter) the spread of fake newscited by 2
  29. 2024Data-Driven Mechanism Design via Multi-Agent Revealed Preferencescited by 2
  30. 2024The Digital Ecosystem of Beliefs: does evolution favour AI over humans?cited by 1
  31. 2025Quantifying the Self-Interest Level of Markov Social Dilemmascited by 1
  32. 2025Game Theory in Social Media: A Stackelberg Model of Collaboration, Conflict, and Algorithmic Incentivescited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Coherence-Driven Belief Formation and Population Dynamics of Contagion in LLM Agents
  2. 2026Physics of Agents: Statistical Mechanics Predicts Collective Behavior of AI Agents
  3. 2026Evaluating Collective Behaviour of Hundreds of LLM Agents
  4. 2026Hidden Anchors in Multi-Agent LLM Deliberation
  5. 2026ScioMind: Cognitively Grounded Multi-Agent Social Simulation with Anchoring-Based Belief Dynamics and Dynamic Profiles
  6. 2026Bayesian Belief Layer for Controllable Opinion Dynamics in LLM Agents
  7. 2026Multi-Agent Consensus as a Cognitive Bias Trigger in Human-AI Interaction
  8. 2026Scaling Behavior of Single LLM-Driven Multi-Agent Systems