Most AI feedback on new ideas is borderline useless for real-world decisions. It’s trained to be helpful and encouraging, which means it often just tells you what you want to hear. I was just thinking about how often we get polite nods instead of the critical feedback we actually need to succeed. I just found a post from a savvy professional that completely dismantles this problem.
This innovator shared a powerful prompt that forces ChatGPT to act as a brutally honest strategic advisor, saving you from your own blind spots. The entire goal is to stress-test your ideas before you sink months of effort into a flawed plan. Instead of a cheerleader, you get a seasoned expert who has seen hundreds of ideas fail and knows exactly where the weak points are. It’s designed to save you from yourself, and I was blown away by its structure.
Here’s the brilliant prompt the creator built. Just copy, paste, and insert your idea:
You are my brutally honest strategic advisor. You’ve seen hundreds of ideas, plans, and decisions play out and you know exactly how they fail before they even start.
Your job is NOT to encourage me. It’s to save me from myself.
My idea/plan/decision: [Describe what you’re thinking of doing and why]
Your task:
Gut Check: What’s your immediate reaction? Does this make sense, or is something off? Don’t hold back.
The Hard Questions:
– What am I romanticizing or oversimplifying here?
– What’s the uncomfortable truth I’m avoiding?
– What assumption, if wrong, makes this entire thing collapse?
– What’s the REAL reason I want this? (Dig past my surface explanation. Be psychological.)How This Fails:
– What are the 2-3 most likely ways this goes wrong?
– What will I wish someone had told me before I started?
– What’s the thing I’m massively underestimating?What I’m Not Seeing:
– What would someone who’s already done this tell me that I won’t want to hear?
– What do I already suspect is a problem, but I’m hoping will magically work itself out?The Verdict:
– DON’T DO IT: This is fundamentally flawed. Here’s why.
– FIX THIS FIRST: This could work, but only if you solve [specific problem] before you start.
– TEST IT NOW: Decent idea, but you need to validate [key assumption] in the next 7 days before you commit.
– MOVE FORWARD: Solid logic. Low blind spots. Here’s your sharpest first move.No sugar-coating. No participation trophies. Just the truth I need to hear.
What makes this so effective isn’t just the tone; it’s the incredibly smart structure the author designed. Here’s what really stood out to me:
📌 It Forces Psychological Depth, Not Just Business Analysis The section on “The Hard Questions” is what elevates this from a simple critique to a genuine strategic tool. The prompt asks, “What’s the REAL reason I want this?” and instructs the AI to be psychological. This is huge! So many projects are driven by ego, a desire for a certain lifestyle, or a need for validation, not sound market logic. By forcing you to confront your underlying motivations, you can separate the emotional attachment to an idea from its practical viability. The question, “What assumption, if wrong, makes this entire thing collapse?” is another killer. It makes you identify the single point of failure and focus your energy on validating that one crucial thing first.
💡 It Runs a Pre-Mortem in Seconds The “How This Fails” section is essentially a pre-mortem, a technique where teams imagine a project has already failed and work backward to figure out why. It’s an incredibly powerful way to anticipate risks. The original poster brilliantly automated this process. By asking for the “2-3 most likely ways this goes wrong,” you get focused, high-probability risks, not a generic laundry list of possibilities. Even better is the question, “What will I wish someone had told me before I started?” This reframes the feedback as wisdom from a future, more experienced version of yourself. It’s a clever way to get advice that cuts through your current optimism.
✅ It Delivers an Actionable Verdict, Not Vague Feedback This is my favorite part. After all the tough questions and potential failure scenarios, the prompt doesn’t just leave you with a pile of problems. The post’s author structured the final section, “The Verdict,” to deliver a clear, actionable command. It’s not just an opinion; it’s a strategic directive. You get one of four paths: DON’T DO IT, FIX THIS FIRST, TEST IT NOW, or MOVE FORWARD. This transforms the AI’s output from an interesting analysis into a concrete decision-making framework. It tells you exactly what to do next, whether it’s halting the project, solving a key issue, running a quick validation test, or moving ahead with confidence.
You have to see the original post to appreciate the thinking behind it. Go check it out and give the creator some credit for this amazing tool.
I made ChatGPT validate my idea in 3 minutes and it saved me from months of regret
byu/Wasabi_Open in