Role-playing vs Self-modelling — LessWrong

In a recent debate on Twitter – which I recommend reading in full – David Chalmers argues: …

x Role-playing vs Self-modelling — LessWrong Language Models (LLMs) LLM Personas AI Frontpage 63 Role-playing vs Self-modelling by Jan_Kulveit 7th Apr 2026 4 min read 3 63 In a recent debate on Twitter – which I recommend reading in full – David Chalmers argues : "Claude doesn't role-play the assistant, it realizes the assistant. Role-playing and realization are quite distinct phenomena, even at the level of behavior and function." Jack Lindsey questions this , pointing out evidence in the opposite direction: "I'm curious what you'd say it's doing when it's sampling tokens on the user turn, or

saved by

• Lydia Nottingham
• Yudhister Joel Kumar
• Eric Huang

related reading

• The Persona Selection Model: Why AI Assistants might Behave like Humansalignment.anthropic.com
• trees are harlequins, words are harlequins - the voidnostalgebraist.tumblr.com
• The persona selection model \ Anthropicanthropic.com
• llm assistant personas seem increasingly incoherent (some subjective observations) — LessWronglesswrong.com
• Simulators — LessWronglesswrong.com
• The persona selection model — LessWronglesswrong.com
• Discovering Language Model Behaviors with Model-Written Evaluations — LessWronglesswrong.com
• Guardian Angels: LLM Personalization for Productivity and Security · Gwern.netgwern.net
• A Mechanistic Explanation of Prompt Injection (and why you should study roles) — LessWronglesswrong.com
• On the functional self of LLMs — LessWronglesswrong.com
• Privilege, Dominance, and Personaseleosai.substack.com
• The Artificial Selftheartificialself.ai