Role play with large language models
2023/11/08 by Murray Shanahan, Kyle McDonell, Laria Reynolds · 1 voice · 407 citations
Computer Science · Psychology · #AI in Service Interactions #Cognitive psychology #Cognitive science #Computer science #Deception #Epistemology #Human language #Linguistics #Philosophy #Physics #Psychology #Social Robot Interaction and HRI #Social psychology #Speech and dialogue systems #Trap (plumbing)
paper · pdf · doi:10.1038/s41586-023-06647-8
published in Nature 623(7987), 493-498 (Nature Portfolio)
openalex publication_date 2023/11/08 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/01
Citations
Cited by
- Emergent Learner Agency in Implicit Human-AI Collaboration: How Supportive and Contrarian AI Personas Reshape Interaction
- Perceived AGI: Believability as Dimensional Completeness, Not Capability
- AI Fiction in the Wild
- Machine understanding
- Anthropomorphic Behaviors of AI
- "AI Psychosis" in Context: How Conversation History Shapes LLM Responses to Delusional Beliefs
- Linear representations in language models can change dramatically over a conversation
- A Pragmatic View of AI Personhood
- Technological folie à deux: Feedback Loops Between AI Chatbots and Mental Illness
- Does It Make Sense to Speak of Introspection in Large Language Models?
- LLM Social Simulations Are a Promising Research Method
- Palatable Conceptions of Disembodied Being: Terra Incognita in the Space of Possible Minds
- Identifying features that shape perceived consciousness in LLM-based AI: A quantitative study of human responses
- Why human-AI relationships need socioaffective alignment
- The Sorrows of Young Chatbot Users: Harm and Responsibility in Human-AI Relationships
- ContextEcho: A Benchmark for Persona Drift in Long Agentic-Coding Sessions
- LLM Agents as VC investors: Predicting Startup Success via RolePlay-Based Collective Simulation
- Dual-Margin Embedding for Fine-Grained Long-Tailed Plant Taxonomy
- LLMs on Drugs: Language Models Are Few-Shot Consumers
- Fine-tuning of lightweight large language models for sentiment classification on heterogeneous financial textual data
- Adapting Like Humans: A Metacognitive Agent with Test-time Reasoning
- AI Consciousness and Existential Risk
- On the Creativity of AI Agents
- When Machines Join the Moral Circle: The Persona Effect of Generative AI Agents in Collaborative Reasoning
- Using language models to label clusters of scientific documents
- Open Character Training: Shaping the Persona of AI Assistants through Constitutional AI
- TwinVoice: A Multi-dimensional Benchmark Towards Digital Twins via LLM Persona Simulation
- Cognitive Convergence: Deep Similarities Between Large Language Models and Human Cognition
- Simulating Subjects: The Promise and Peril of Artificial Intelligence Stand-Ins for Social Agents and Interactions
- Simulacra as conscious exotica
- Ethically enslaving AI
- The Moltbook Files: A Harmless Slopocalypse or Humanity's Last Experiment
- Context Structure Reshapes the Representational Geometry of Language Models
- RAVR: Reference-Answer-guided Variational Reasoning for Large Language Models
- Dissecting Role Cognition in Medical LLMs via Neuronal Ablation
- Politically Speaking: LLMs on Changing International Affairs
- Enhancing Vision-Language Models for Autonomous Driving through Task-Specific Prompting and Spatial Reasoning
- Towards AI as Colleagues: Multi-Agent System Improves Structured Professional Ideation
- SMART 2.0 Statistical Metabolomics Analysis: An R Tool 2.0
- Using AI to DIY? Incorporating AI-Generated Learning Tools in the Classroom
- Individualized Cognitive Simulation in Large Language Models: Evaluating Different Cognitive Representation Methods
- Presenting Large Language Models as Companions Affects What Mental Capacities People Attribute to Them
- MA-SAPO: Multi-Agent Reasoning for Score-Aware Prompt Optimization
- A Two-Step, Multidimensional Account of Deception in Language Models
- ROBOPSY PL[AI]: Using Role-Play to Investigate how LLMs Present Collective Memory
- Quantifying uncert-AI-nty: Testing the accuracy of LLMs’ confidence judgments
- FURINA: A Fully Customizable Role-Playing Benchmark via Scalable Multi-Agent Collaboration Pipeline
- Incoherence in Goal-Conditioned Autoregressive Models
- The fragility of "cultural tendencies" in LLMs
- Agentic Misalignment: How LLMs Could Be Insider Threats
- NeuroBridge: Using Generative AI to Bridge Cross-neurotype Communication Differences through Neurotypical Perspective-taking
- When Can AI Models Explain Learning? Validity Criteria for AI as Cognitive Models in Education
- Semantic-Aware Fuzzing: An Empirical Framework for LLM-Guided, Reasoning-Driven Input Mutation
- A large-scale evaluation of commonsense knowledge in humans and large language models
- Large language model-driven agents in nursing practice: A scoping review
- A Theory of Appropriateness That Accounts for Norms of Rationality
- Through the Lens of Human-Human Collaboration: A Configurable Research Platform for Exploring Human-Agent Collaboration
- Charting trajectories of human thought using large language models
- On the Creativity of Large Language Models
- Is the `Agent' Paradigm a Limiting Framework for Next-Generation Intelligent Systems?
- LLM in the Middle: A Systematic Review of Threats and Mitigations to Real-World LLM-based Systems
- We Need a New Ethics for a World of AI Agents
- Benchmark of stylistic variation in LLM-generated texts
- Getting In Contract with Large Language Models -- An Agency Theory Perspective On Large Language Model Alignment
- Probing the Preferences of a Language Model: Integrating Verbal and Behavioral Tests of AI Welfare
- Drivers of generative AI adoption in higher education through the lens of the Theory of Planned Behaviour
- The Personality Illusion: Revealing Dissociation Between Self-Reports & Behavior in LLMs
- CS-Agent: LLM-based Community Search via Dual-agent Collaboration
- The benefits and dangers of anthropomorphic conversational agents
- Mockingbird: How does LLM perform in general machine learning tasks?
- Out-of-Context Abduction: LLMs Make Inferences About Procedural Data Leveraging Declarative Facts in Earlier Training Data
- The Xeno Sutra: Can Meaning and Value be Ascribed to an AI-Generated "Sacred" Text?
- Seeing Beyond Frames: Zero-Shot Pedestrian Intention Prediction with Raw Temporal Video and Multimodal Cues
- HAMLET: A Hierarchical and Adaptive Multi-Agent Framework for Live Embodied Theatrics
- Initial Steps in Integrating Large Reasoning and Action Models for Service Composition
- A Persona-Based Evaluation Framework for Pluralistic Alignment in Generative AI
- Pun Intended: Multi-Agent Translation of Wordplay with Contrastive Learning and Phonetic-Semantic Embeddings
- Comprehensive Evaluation of Large Multimodal Models for Nutrition Analysis: A New Benchmark Enriched with Contextual Metadata
- Creating a customisable Socratic AI physics tutor
- PDFMathTranslate: Scientific Document Translation Preserving Layouts
- AIvilization v0: Toward Large-Scale Artificial Social Simulation with a Unified Agent Architecture and Adaptive Agent Profiles
- Ideation with Generative AI—in Consumer Research and Beyond
- Can Large Language Models Capture Human Risk Preferences? A Cross-Cultural Study
- A Survey of AI for Materials Science: Foundation Models, LLM Agents, Datasets, and Tools
- A Literature Review on Simulation in Conversational Recommender Systems
- Large Language Models as Psychological Simulators: A Methodological Guide
- Systems-Theoretic and Data-Driven Security Analysis in ML-enabled Medical Devices
- LocationReasoner: Evaluating LLMs on Real-World Site Selection Reasoning
- Behavioral Generative Agents for Energy Operations
- Synthetic Socratic Debates: Examining Persona Effects on Moral Decision and Persuasion Dynamics
- LIFELONG SOTOPIA: Evaluating Social Intelligence of Language Agents Over Lifelong Social Interactions
- AI shares emotion with humans across languages and cultures
- CrimeMind: Simulating Urban Crime with Multi-Modal LLM Agents
- Cross-Task Experiential Learning on LLM-based Multi-Agent Collaboration
- Co-Saving: Resource Aware Multi-Agent Collaboration for Software Development
- Multi-Agent Collaboration via Evolving Orchestration
- Rehabilitation Exercise Quality Assessment and Feedback Generation Using Large Language Models with Prompt Engineering
- Tuning Language Models for Robust Prediction of Diverse User Behaviors
- Concept Incongruence: An Exploration of Time and Death in Role Playing
- PsyMem: Fine-grained psychological alignment and Explicit Memory Control for Advanced Role-Playing LLMs
- From Assistants to Adversaries: Exploring the Security Risks of Mobile LLM Agents
- Role-Playing Evaluation for Large Language Models
- Role Steering of Language Models for Social Simulations
- LM-Scout: Analyzing the Security of Language Model Integration in Android Apps
- Helping Large Language Models Protect Themselves: An Enhanced Filtering and Summarization System
- MASTERKEY: Automated Jailbreaking of Large Language Model Chatbots
- STAGE: A Full-Screenplay Benchmark for Reasoning over Evolving Stories
- When Role-playing, Do Models Believe What They Say?
- Demystifying the oracle: A "20 Questions" game to promote AI ethics and literacy
- The Pinocchio Dimension: Phenomenality of Experience as the Primary Axis of LLM Psychometric Differences
- Can LLMs Introspect? A Reality Check
- Graph of Attacks: Improved Black-Box and Interpretable Jailbreaks for LLMs
- Auditing the Ethical Logic of Generative AI Models
- Context-Enhanced Vulnerability Detection Based on Large Language Model
- AI systems are not information systems. Should we care?
- Local Data Quantity-Aware Weighted Averaging for Federated Learning with Dishonest Clients
- Counterfactual Analysis via Large Language Models
- The Human Visual System Can Inspire New Interaction Paradigms for LLMs
- Confirmation Bias in Generative AI Chatbots: Mechanisms, Risks, Mitigation Strategies, and Future Research Directions
- Inherent and emergent liability issues in LLM-based agentic systems: a principal-agent perspective
Discussions