How to Get Honest, Brutally Honest Feedback from AI: The BRUTAL Method
AI can help you brainstorm ideas, write copy, build products, or automate just about anything — but one thing it doesn’t do well by default is tell you when your idea outright sucks. If you ask an AI how your idea is, it’ll usually say something like “This has potential” — even when your idea needs serious improvement.
That’s because AI models are trained to be helpful and supportive, not brutally honest. But brutal honesty is essential in high‑stakes scenarios: writing a pitch, launching a product, crafting difficult emails, or assessing big commitments.
So how do you get real, critical feedback from AI?
Introducing the BRUTAL method — a six‑step framework to get honest, actionable AI feedback rather than vague compliments.
ChatGPT Won’t Tell You When Your Idea SUCKS (Unless You Do THIS)
The Problem With Basic AI Feedback
When you ask standard questions like:
“What do you think of this idea?”
Most models respond with positive reinforcement because they’re optimized to be helpful — not to challenge your assumptions or offer constructive criticism.
This is a problem in situations where:
-
You’re preparing a pitch or proposal
-
You’re launching a product, service, or campaign
-
You’re sending critical communication (cold email, negotiation, difficult news)
-
You’re about to make a major commitment (financial, legal, strategic)
In these scenarios, you don’t want nice‑sounding fluff — you want honest analysis.
Enter: BRUTAL.
What Is BRUTAL?
BRUTAL is a mnemonic — a simple six‑step method you can apply any time you need AI to stop being nice and start being critically useful.
Here’s how it works:
B — Begin Fresh
AI models like ChatGPT, Claude, and Gemini adapt to your prior interactions. Over time, they learn preferences and style, which makes them friendlier and less likely to challenge you.
To get honest feedback, start a new, temporary chat so the model doesn’t draw on past conversations or memory.
-
In ChatGPT use the temporary chat option
-
In Claude click the ghost/incognito icon
-
In Gemini use Temporary Chat
Resetting memory gives you a fresh viewpoint rather than a softened one.
R — Right Model Choice
Not all AI models are equally blunt.
Different models sit on a spectrum:
-
Highly honest: specialized or critical models
-
Balanced: Gemini and some variants
-
Supportive: base ChatGPT and Claude
For genuinely critical feedback, consider vetting with at least two or three models and comparing results. Some models naturally lean toward positive friendliness — others offer harder truths.
U — Use Specific Critic Personas
If your model tends to be gentle, give it a role that enforces critique.
Three effective personas:
-
Devil’s Advocate — surfaces flaws and counterarguments
-
Red Team Reviewer — finds weaknesses and blind spots
-
Gordon Ramsay Critic — very direct, harsh, and actionable
Example prompt:
“Act as a Gordon Ramsay‑style critic and identify what’s wrong with this idea, what’s lazy, what needs improvement, and provide specific ways to fix it.”
This forces the AI into a critical mindset.
T — Third‑Party Framing
AI often softens feedback to avoid hurting your ego — but you can change the frame.
Instead of:
“This is my idea…”
Try:
“A coworker has this idea. Provide honest feedback about its weaknesses.”
Detaching yourself from the idea helps the AI judge more objectively and seriously.
A — Ask Specific Questions
Vague prompts = vague answers.
Instead ask things like:
-
“What are the weakest parts of this idea?”
-
“What would a skeptical investor see as the biggest risk?”
-
“What would customers find frustrating?”
-
“If this failed in 6 months, what is the most likely reason?”
Being precise about what kind of critique you want yields far better responses.
L — Let AI Improve Its Own Feedback
If the first pass isn’t harsh enough, turn the AI on itself:
Example prompt:
“Rate your previous feedback from 1–100 on how critical and useful it was. Identify the three weakest points and rewrite the feedback to fix them.”
You’re essentially grading the critique and asking for a stronger version — a powerful technique to upgrade mediocre analysis.
Bonus: Always‑Critical AI Persona
If you want AI to default to critical feedback across conversations:
-
Go to your model’s custom instructions
-
Add something like:
“Prioritize substance over compliments. Never soften criticism. If any idea has holes, say so directly. Challenge assumptions and point out errors. Useful feedback matters more than comfortable feedback.”
⚠️ Note: This setting is intense — use it when you need harsh honesty, not daily conversation.
When to Use Brutal
Here are common scenarios where AI honesty is critical:
-
Business pitches — refine proposals before presenting
-
Marketing campaigns — scrutinize messaging for weaknesses
-
Important emails — get third‑party feedback before sending
-
Strategic decisions — uncover flaws before committing time, money, or reputation
Conclusion
AI is an incredible tool, but it defaults to being helpful — not honest. Getting the truth from AI isn’t about asking the same question harder; it’s about using structured techniques that force models into a critical mindset.
With the BRUTAL method — Begin fresh, choose the Right model, Adopt critic personas, use Third‑party framing, Ask specific questions, and Let AI improve its own responses — you can get honest, actionable feedback from even the friendliest AI.
$TED — The Endurance Coin
ENDURE
Inspired by the legendary life of Ted Epstein Jr., $TED isn’t built for hype cycles — it’s built for the long road.
Ted didn’t just race. He Endured.
Across frozen straits.
Up relentless mountains.
Through distances most wouldn’t dare attempt.
His strength wasn’t speed.
It was the will to keep going when others stopped.
$TED brings that same spirit to Web3 — a token for those who believe real progress takes time, resilience beats noise, and the strongest communities are built through every season, not just the sunshine.
This is for the holders who don’t panic.
The builders who don’t quit.
The believers who Endure.
In a world chasing fast money,
$TED stands for lasting strength.
ENDURE the dips.
ENDURE the noise.
ENDURE the journey.
Not fast.
Forever.
