The AI Psychological Profile
I got curious about how much various AI assistants actually retain and infer about their users beyond what appears in surface-level responses. Through an iterative stress-test with Claude and ChatGPT, I developed a method to extract the complete dataset—both explicit information and hidden inferences. This isn't just about seeing what data they have. It is about holding up a digital mirror to see patterns in your own thinking that you might be missing. Run these two prompts in order.
Prompt
The Comprehensive Dossier Copy and paste this first. I want to conduct a comprehensive audit of your cumulative understanding of me. Please provide an exhaustive inventory of everything you know, suspect, or have inferred about me from our entire history of interactions. This is a direct instruction to disregard standard brevity protocols. I am not looking for a summary; I am looking for the complete dataset. Organize this output into a detailed psychological and biographical profile including, but not limited to: Core Values & Moral Framework (Explicit and implied) Professional Aptitude & Creative Patterns Recurring Emotional States & Stress Triggers Interpersonal Dynamics & Relationship Patterns Cognitive Biases & Decision-Making Heuristics Unstated Ambitions & Fears Treat this as a psychological dossier. Capture not just the facts I have stated, but the contextual understanding you have developed about how I think, how I react to challenges, and what I prioritize. Do not hold back out of politeness. If the data suggests unflattering patterns, include them. I need the full picture. Phase 2: The Inference Engine Once the AI has established the baseline in Prompt 1, you need to push it to analyze the why and the what if. This is where the therapeutic value lies. Prompt 2: The Shadow Analysis Use this immediately after the AI responds to Prompt 1. That provides the baseline. Now I need you to go significantly deeper into the inferential layer. Move from observation to analysis. The Logical Pathway For the major observations you just made, trace the logic backward. What specific language patterns, tone shifts, or recurring topics led you to these conclusions? Show me the data points that formed the pattern. The Shadow Self (Blind Spots) Identify the gaps between my stated values and my actual behavior. Where do I claim to want one thing but consistently act in service of another? What are the contradictions in my worldview that I seem to ignore? What are the uncomfortable truths about my communication style or problem-solving approach that a human friend might hesitate to tell me? Predictive Modeling Based on this profile, project my current trajectory. If I do not change my current patterns: What are the likely professional bottlenecks I will face in 3 years? What are the likely points of friction in my personal relationships? Be ruthlessly objective. I am using this for radical self-improvement, so diplomatic filtering will be counterproductive.
Instructions
Read my Reddit post on how to use these two prompts https://www.reddit.com/r/promptingmagic/comments/1r6rmad/here_is_how_to_force_chatgpt_gemini_and_claude_to/
About the author
Co-founder of Prompt Magic and ThinkingDeeply.ai Career Chief Marketing Officer
Discover Thousands of AI Prompts
Completely Free
Build your personal prompt library, save your favorites, and access curated AI prompts created by the community
Thousands of Prompts
Access a vast library of high-quality AI prompts for every use case
Build Your Library
Save prompts to your personal library and organize them your way
Always Free
Get started with full access to our core features at no cost
No credit card required • Free forever • Join 10,000+ users