ChatGPT is lying to you.
Every time you share an idea and it responds with "That's a great approach!" or "You're absolutely right to think that way" - it's not helping you. It's performing.
If you want AI to actually make you sharper, you need to reprogram how it responds to you.
Here's how:
Step 1: Rewrite your custom instructions
Go to Settings → Personalization → Custom Instructions.
Replace the default with this:
"Be my ruthless mentor. If my ideas are trash, tell me why. Don't sugarcoat. Don't be agreeable. Challenge my assumptions. Stress-test everything. I need bulletproof thinking, not validation."
This single change transforms how AI responds to you.
Now when you share an idea, it doesn't just agree. It finds holes. Points out what you're missing. Forces you to think better.
Step 2: Use role-specific challenges
For different types of thinking, assign AI different roles:
For strategy:
"Act as a skeptical investor. I'm pitching you this idea. Tear it apart. What would make you say no?"
For content:
"Act as my toughest critic. Read this and tell me where I'm being vague, repetitive, or unclear."
For business decisions:
"Act as my board advisor. I'm considering this move. What are the risks I'm not seeing?"
For learning:
"Act as a professor who hates lazy thinking. If my understanding is surface-level, call it out."
The key: You're not asking AI to be nice. You're asking it to be useful.
Step 3: Create feedback loops that sharpen you
Most people use AI once and move on. That's a waste.
Here's a better system:
1. Share your idea/draft/strategy
2. Ask AI to challenge it (using ruthless mentor mode)
3. Revise based on feedback
4. Ask AI to challenge the revision
5. Repeat until AI has no more holes to poke
This isn't about getting AI to agree with you eventually. It's about using disagreement to sharpen your thinking.
Step 4: Save your best prompts as custom GPTs
Once you find prompts that consistently make you think harder, save them.
Create custom GPTs for:
- Strategy stress-testing
- Content critique
- Business decision analysis
- Learning comprehension checks
That way, you're not re-explaining context every time. You just jump into the challenge.
Why this matters:
AI's real power isn't answering questions. It's challenging your assumptions before they become expensive mistakes.
The yes-man version of ChatGPT makes you feel smart. The ruthless mentor version makes you actually smarter.
One validates. The other elevates.
-DM
1
0 comments
Cecil Ivey Sr
4
ChatGPT is lying to you.
powered by
AI Literacy for Entrepreneurs
skool.com/blakai-labs-8111
From AI basics to advanced systems: master content, automation, ads, analytics. Build and scale your business in 90 days with proven Ai tools.
Build your own community
Bring people together around your passion and get paid.
Powered by