AI Won't Tell You Your Idea Sucks #ai #verification #prompt #brutal #honesty #devil #advocate #brutal #pattern #sucks #critical #feedback

ai_brutal.md

BRUTAL

B - Begin Fresh R - Right Model U - Use a Critic Persona T - Third Party Framing A - Ask Specific Questions L - Let It Grade Itself

Honest and Critical Feedback

The problem is AI wants to please you. It will return something like: "This has potential! Here's how to make it even better..."

Moments That Need Brutal Honesty

• Before you pitch a proposal to a client or boss.
• Before you send a negotiation, cold pitch, or difficult news.
• Before you launch a product, service, or campaign.
• Before you commit to contracts, vendors, or major expenses.

Memory Makes AI Agree With You

• Begin Fresh - Start with a clean slate
  • Ralph Wiggum
  • ChatGPT
    • Temporary chat
    • Blocks long-term memory
    • Custom instructions still apply
  • Claude
    • Incognito mode
    • Click the ghost icon
    • Bypasses all saved memory
  • Gemini
    • Temporary Chat
    • Personal use toggle
    • Work already private
• AI Memory
  • You chat more
  • It learns patterns
  • Less honest feedback
  • It adapts to please

Use A Critic Persona

• Devils Advocate
  • Challenge with counterarguments
• Red team
  • Hunt for weaknesses on purpose
• Gordon Ramsay
  • Surgical
  • No sugarcoating

Devil's Advocate (Good Daily Driver)

Act as a devil's advocate. Challenge this idea by pointing out flaws, counterarguments, miussing evidence, and unintended consequences. Your goal is to create doubt.

Red Teaming (Great for Proposals, contracts, business plans, product ideas)

Act as a red team reviewer. Your job is to find and poke holes in (X)--expose flaws, loopholes, and things that arent good enough by purposefully hunting for them.

Gordon Ramsay Treatment

Give this the Gordon Ramsay treatment. Be surgical. What's wrong, what's lazy, and what needs to be completely redone? Make sure the feedback is specific and actionable.

Third Party Framing (Same idea, different framing/Your idea vs Coworker's idea)

AI doesn't try to protect your ego.

A coworker has this idea and asked for honest feedback. Help identify the weaknesses so they don't waste time on something flawed. Here's the idea: (X).

A random person asked for feedback on their business plan.

Someone on my team proposed...

A founder I'm advising is considering...

Specific Questions Unlock Specific Criticism

1. What's the weakest part that's easy to miss?
2. What would a skeptical investor say is the biggest risk?
3. What might customers find frustrating about this?
4. If this fails in siz months, what's the likely reason?

Let It Grade Itself

After all those, AI might still hold back criticism Ask AI to grade itself. After self-critique its much sharper. AI switches from "be helpful" mode to "find problems" mode.

Analyze that response. Rate it from 1 to 100 for how genuinely critical and useful it was. Identify the three weakest points, then rewrite the feedback fixing those weaknesses.

Set It Once And Forget It

Settings -> Personalization -> Custom Instructions

Prioritize substance over compliments. Never soften creiticism. If an idea has holes, say so directly--"This won't scale becauseX" is better than "Have you considered...". Challenge assumptions. Point out errors. Useful feedback matters more than comfortable feedback.