Overview
How might we design persona selector so people can trust and act on AI output?
When this pattern fits
- Perfect for multi-purpose AI tools, conversational applications, and platforms where switching AI personas adapts the assistant to different tasks and contexts.
When to skip or lighten it
- Products with one fixed professional voice and no role diversity.
- Safety-critical advice where role-play could imply unqualified expertise.
- Crowded composers where persona chips compete with mode and model controls.
States
State model coming soon
Key UX elements
Key UX elements coming soon
Anti-patterns to avoid
Personas that claim credentials the product cannot guarantee.
Switching persona mid-thread without showing what instructions changed.
Dozens of near-duplicate personas with no search or categories.
Persona names that do not match the instructions users actually get.
How products use it
| Product | Implementation |
|---|---|
| ChatGPT | GPTs and role-oriented custom instructions selectable per chat. |
| Character.ai | Character grid as the primary persona selection surface. |
| Claude | Styles and project-level instruction personas for work contexts. |
| Perplexity | Focus/profile modes that bias research toward a domain role. |
Real-world examples
How shipped products implement persona selector, from our teardown guides.
ChatGPTChatGPT composer UX: input bar, tools & voice design
Design the composer
ChatGPTChatGPT personalization UX: tone presets, memory & profile design
Personalization & memory
ClaudeClaude composer UX: model picker, skills & attachments
Design the composer
ClaudeClaude personalization UX: profile, memory & voice design
Personalization & memory
GeminiGemini composer UX: tools, Drive & thinking modes
Design the composer
LovableLovable composer UX: Build, Plan & templates
Design the composer
PerplexityPerplexity composer UX: search modes & connectors
Design the composer
PerplexityPerplexity personalization UX: vertical profiles, response style & memory
Personalization & memory
Implementation
Copy this prompt to generate a production-ready implementation in Cursor, Claude Code, Lovable, or any AI coding agent.
Generate a production-ready implementation of the "Persona Selector" AI interface design pattern.
Pattern Definition:Frequently asked questions
What is a persona selector in AI UX?
A persona selector lets users pick a named assistant role (such as tutor or editor) that applies a preset instruction set and expectations for the conversation.
Persona vs model selection?
Model selection chooses capability and cost. Persona selection chooses behavior and framing on top of a model. Users often need both, labeled separately.
Should personas include example prompts?
Yes. Pairing a persona with two or three starter prompts teaches how to talk to that role and reduces blank-composer friction.
Can personas override safety policy?
No. Product policy must win. Personas can change tone and task framing, not permissions to give dangerous or disallowed help.