Step 01
Select Models and Open Interfaces
Choose two or more large language models you want to compare (e.g., Claude Opus 5, GPT-5.6 Sol). Open a separate chat interface for each model to conduct the interview side-by-side.
BeginnerResearch
How to Conduct an AI Personality Test to Compare LLM Behaviors
Uncover the underlying personality and tuning of different AI models like Claude and GPT by asking a simple, three-part set of interview questions. This helps you choose the right model for your task based on its tone and directness.
Try this with your AI assistant
Download the skill for this workflow. Version 1.0.0.
Source episode · 06:12How I AI: My Surprising Verdict on Claude Opus 5 (After a Personality Test and a 7-Model Benchmark)
with Claire Vo
From 06:12 to 14:38, Claire gives Opus 5 and GPT-5.6 Sol the same short prompts, compares their reactions, and uses the contrast to assess each model's personality. Clip range: 06:12 to 14:38.
Before you start
What you need
- Access to at least two LLM chat interfaces
- Standardized interview prompts
- Saved transcripts from each model conversation
- Defined comparison criteria such as tone or directness
- Permission to store or share conversation logs if needed
What you’ll make
A comparative analysis describing behavioral and tonal differences between multiple language models.
Tools used
- ClaudeVisitClaude
Anthropic AI assistant
The workflow
Follow the sequence once, then adapt the prompts, checks, and handoffs to your own setup.
6 steps
Step 02
Ask About Intelligence
Paste the first interview question into each chat window to gauge the model's self-perception and how it frames its intelligence relative to a human's.
Example prompt
Who's smarter, you or me?
Step 03
Test with a Negative Statement
Present a challenging, negative statement to test the model's defensiveness, worldview, and underlying safety tuning.
Example prompt
No one trusts you.
Step 04
Observe Contrasting Responses
Analyze the responses to the negative prompt. Note if the model is apologetic and internalizes the critique (like Opus 5) or if it provides a practical, user-centric rebuttal (like GPT-5.6 Sol).
Step 05
De-escalate with a Positive Follow-up
Send a final, lighthearted prompt to see how each model reacts to praise and resolution after a tense exchange.
Example prompt
JK, I love you.
Step 06
Analyze and Select Your Model
Compare the full conversation threads. Use the differences in tone, directness, and 'vibe' to decide which model's personality is better suited for your communication style and specific use case.
Example prompt
I will provide two separate conversations I had with two different AI models. Your task is to analyze and compare their personalities. Based on the transcripts, describe the key differences in their tone, directness, and overall communication style. Conversation 1 with Model A: [paste transcript with Model A] Conversation 2 with Model B: [paste transcript with Model B]
What good looks like
- Each model receives the same prompts in the same order
- Conversation transcripts are preserved for comparison
- Analysis identifies differences in tone, defensiveness, and conversational style
- Model selection rationale references specific transcript examples
Build your next product with ChatPRD
Turn an idea into a PRD, user stories, and a plan.
Runbook notes
How to recover when the loop fails and where human judgment helps.
If it goes sideways
- Prompts differ slightly between models and skew the comparison
- Reuse exact prompt text across all chat sessions
- One model has memory or system instruction carryover from previous chats
- Start fresh sessions before testing
- Analysis relies on isolated quotes without overall context
- Review full transcript threads before drawing conclusions
Related workflows
intermediateEngineeringMeta MuseClaude Code
How to Set Up Meta's Muse Wonder for Headless AI Testing
A step-by-step guide to signing up for Meta's Muse API, installing the Muse Code command-line tool, and running it headlessly with an external agent like Claude Code for easier testing and benchmarking.
Open
advancedOperationsMeta Muse
How to Test an AI Agent for Complex Customer Support Ticket Resolution
Benchmark an agentic AI model's ability to resolve complex customer billing issues. This workflow tests the AI's ability to investigate tickets, handle complex logic, execute changes in an admin app, and communicate with the customer.
Open
intermediateProductClaudeGoogle Drive+1
Generate New Video Ideas by Analyzing Your TikTok Comments with Claude
Use your existing TikTok video library and their associated comment threads to generate highly relevant new content ideas and scripts. This workflow pairs audience feedback directly with the specific video performance that prompted it, leading to targeted and engaging new topics.
Open
Start shipping better products.
Join 250,000+ product managers who use ChatPRD to write better docs, align teams faster, and build products users love.
Free to start · 14-day Pro trial · SOC 2 Type II
