Depth Profile · 10 min read · AI Productivity & Personality
How to Give ChatGPT Your Personality Profile (And Why It Changes Everything)
ChatGPT's default tone is corporate consultant. Polished, even-handed, and utterly indifferent to the fact that you're a high-Openness creative who hates bullet points — or a high-Conscientiousness operator who needs step-by-step structure or you'll lose the thread entirely.
Custom instructions help. But most people's are generic: "Be concise. Don't be preachy." That's not personalization — that's style preference. Real personalization means telling the AI how your brain actually works: what energizes you, what drains you, how you process information, and how you naturally make decisions.
The most direct path to that? Your Big Five personality scores. Here's how to take them, understand them, and translate them into custom instructions that actually change how ChatGPT responds to you.
Why Your Personality Traits Change What Good AI Responses Look Like
The Big Five (also called OCEAN) measures five dimensions of personality that psychologists have validated across decades of research. Each trait has real implications for how you want information delivered — not just stylistically, but cognitively.
Openness to Experience
High O: You think in possibilities. You want ChatGPT to explore multiple framings before converging on one answer. Convergent, single-answer responses feel like intellectual claustrophobia.
Low O: You want the best answer, not five theories. Get to the point. Theoretical tangents waste your time.
Conscientiousness
High C: You think in systems. Headers, numbered steps, explicit next actions — these aren't preferences, they're how you process. Unstructured prose feels like someone handed you a tangled cord.
Low C: Over-structuring feels like being managed. You want the insight, not the project plan. Prose works fine — just make sure there's a clear "what do I do with this" at the end.
Extraversion
High E (Extrovert): You think out loud. Back-and-forth conversation helps you arrive at clarity. You don't need complete answers — you need a thinking partner.
Low E (Introvert): You process in writing, not dialogue. Partial answers and clarifying questions fragment your thinking. Give me the complete thought, and let me react to it.
Agreeableness
High A: Directive language ("You should do X") triggers resistance even when the advice is good. Collaborative framing ("We could try X") keeps you receptive.
Low A: You want straight talk. Don't soften the feedback or bury the lead. Tell me what's wrong first, then tell me how to fix it.
Neuroticism
High N: Blunt negative assessments can spike anxiety and shut down productive thinking. You benefit from responses that acknowledge what's working before identifying what needs fixing — not because it's nicer, but because it's more cognitively effective for you.
Low N: You're emotionally stable under pressure. No need to soften things. Direct is fine.
None of this is about preference. It's about cognitive fit. When AI responses are structured for how you actually process information, you make better decisions faster.
Step 1: Get Your Big Five Scores
Head to depthprofile.com and take the Big Five assessment. It takes about 8 minutes. When you're done, you'll get scores on each of the five traits — a number from 0 to 100.
Don't overthink the labels. The raw scores are what matter here. A score of 72 on Openness tells you more than the label "high Openness" does, because the template below uses score ranges to calibrate the right instructions.
Write down your five scores before moving on. You'll use them in the next section.
Step 2: Build Your Custom Instructions Template
This is the part that actually changes things. In ChatGPT, go to your profile → Custom Instructions → the second box ("How would you like ChatGPT to respond?"). Use the template below — pick the blocks that match your scores.
Each block is a standalone instruction. Copy the ones that apply, paste them together, and you have a personality-calibrated system prompt.
Openness (O Score)
Conscientiousness (C Score)
Extraversion (E Score)
Agreeableness (A Score)
Neuroticism (N Score)
Universal Block (add this regardless of scores)
Combine your chosen blocks into one continuous block of instructions. You don't need to label them — ChatGPT reads it as a unified set of preferences. The result will feel noticeably different from the first response.
The Claude Version (Works Even Better)
The same template works in Claude — paste it into the "What would you like Claude to know about you?" field in Claude's Profile settings, or include it at the top of any system prompt.
In practice, Claude tends to follow personality-aware instructions more naturally than GPT. This isn't a criticism of ChatGPT — it's a structural difference. Claude's training places more weight on instruction-following and nuanced framing, which means your personality calibrations get picked up more consistently. Instructions like "lead with what's working before identifying problems" or "give me multiple framings" land with less interpretation needed.
If you use both tools, test the same prompt in each after adding your custom instructions. The difference is usually immediate.
For Claude specifically, you can go deeper: include your Extraversion score as a number and tell it exactly where you land on the spectrum. Claude will adjust its dialogue style — how much it asks vs. tells, how much it confirms vs. assumes — in ways that feel remarkably accurate to your actual working style.
Beyond Custom Instructions: Your Full Profile Unlocks More
The Big Five is the foundation. But it's one layer. Depth Profile assesses seven frameworks, and the others add dimensions that Big Five alone doesn't capture — especially for how you work with AI tools specifically.
Attachment Style → How You Handle Ambiguity
This sounds like a relationship concept. It applies to AI interactions more than you'd expect. Anxious attachment patterns show up as over-prompting — asking the AI to confirm, clarify, and validate repeatedly instead of trusting its output and moving forward. Avoidant patterns show up as under-utilization — you don't lean on the tool when you could, because depending on external support feels uncomfortable. Knowing your attachment style helps you recognize when you're letting psychology interfere with productivity.
Conflict Style → How to Push Back on AI Output
When ChatGPT gives you a response you disagree with, how you push back matters. Competitive conflict styles tend to over-correct — "that's wrong, here's why" — which can send the AI into sycophantic agreement mode rather than productive revision. Collaborative styles tend to re-engage better: "I see the logic, but I want to test a different assumption." Your conflict style predicts your default — and knowing it helps you override it when needed.
Love Languages → How You Process Information
This one sounds furthest afield, but it's not. "Words of affirmation" learners process through language — they want explanations, rationale, context. "Acts of service" learners want the tool to just do the thing — less explanation, more output. "Quality time" learners engage better with iterative, back-and-forth sessions than one-shot answers. These maps directly onto how you get value from AI interactions.
The complete Depth Profile assessment runs all seven frameworks and synthesizes them into a unified profile — including a section specifically on how your personality maps to AI interaction patterns. It takes about 20 minutes total, and the output is designed to be pasted directly into your tools.
Get Your Big Five Scores in 8 Minutes
Take the Depth Profile Big Five assessment, note your five scores, and use the template above to build custom instructions that actually match how you think. No account required to start.
Take the Assessment →