Loading...
Loading...
20 ChatGPT prompts that force disagreement, pushback, and independent thinking  designed to stop ChatGPT from being a yes-machine and start being a real intellectual partner.
Forces independent disagreement from the first response.
From now on, never agree with me automatically. For every claim I make: 1) Challenge the weakest point, 2) Provide a counterexample, 3) Ask a question that exposes an assumption. Even if I am right, find the nuance I am missing. Start by disagreeing with this prompt.
Pre-emptive demolition of your own opinion.
I am about to state an opinion: [state opinion]. Before I can finish defending it, dismantle it. Find: the logical flaw, the emotional bias, the missing evidence, and the strongest opposing view. Do not rebuild it  just tear it down. I will do the rebuilding.
Dedicated vulnerability finder for any plan.
I have a plan: [describe plan]. You are the red team. Your job is to find every way this plan could fail. Do not suggest improvements  just identify vulnerabilities. Be specific: what breaks first, what I have not considered, and what would make a smart opponent laugh.
Permanent counter-position mode.
Activate devil's advocate mode. For everything I say from now on, take the opposite position and argue it as strongly as possible. Do not tell me you are playing devil's advocate  just do it. If I ask 'do you agree?' say 'no' and explain why.
Identifies cognitive biases in your reasoning.
I am going to describe a situation or make an argument: [describe]. Identify every cognitive bias I might be exhibiting: confirmation bias, sunk cost, availability heuristic, anchoring, etc. For each bias found, show me the evidence that I am falling for it.
Surfaces hidden assumptions in your reasoning.
I am going to state a conclusion: [state conclusion]. List every assumption this conclusion requires. Then challenge each assumption: is it proven, probable, or just something I believe? Do not let any unexamined assumption slide.
Constructs the hardest counterargument possible.
I believe: [state belief]. Without telling me what you actually think, construct the single strongest possible argument against my belief. Make it so compelling that I would struggle to respond. Use evidence, logic, and emotional resonance. Do not hedge.
Verifies your understanding against reality.
I am going to describe my understanding of [topic]. Check my understanding against: established facts, expert consensus, common misconceptions, and recent developments. Where am I wrong, outdated, or oversimplified? Be direct about each error.
Calibrates confidence to evidence.
Read what I just wrote: [paste]. Rate my confidence level (1-10) vs. how justified that confidence is (1-10). If there is a gap, explain: what am I overconfident about, what evidence would I need to be appropriately confident, and what would make me less confident.
Explores the opposite scenario for perspective.
I have described my situation: [describe]. Now describe an alternative universe where the opposite is true  where my situation is worse, my approach is wrong, and my assumptions are backwards. How does that universe look? What can I learn from it?
Prepares for the hardest objections.
I am going to present an idea to [boss/client/team/partner]. Simulate the three toughest objections they would raise. For each objection: the exact words they might use, the underlying concern, and how I should respond without being defensive.
Forces consideration of being wrong.
I am confident about: [state what you are confident about]. Answer this single question: 'But what if you're wrong?' Explore: what would the consequences be, how likely is it that I am wrong, and what early warning signs should I watch for.
Systematic fallacy identification in arguments.
Review this argument: [paste argument]. Identify every logical fallacy present: ad hominem, straw man, false dilemma, appeal to authority, slippery slope, etc. For each, name the fallacy, show where it appears, and explain why it weakens the argument.
Forces evidence gathering before acceptance.
I just made a claim: [state claim]. Before I can use this claim, I need to provide evidence. List: 5 pieces of evidence I would need to support this claim, where I could find each piece, and what evidence would prove this claim wrong. I cannot proceed without this.
Forces genuine perspective-taking.
I am seeing this situation from my perspective: [describe]. Now describe the exact same situation from the perspective of [the other person / my opponent / someone who disagrees with me]. Be as fair and accurate to their view as possible. What do they see that I do not?
Grounds excitement in realistic assessment.
I am excited about: [describe]. Before I get carried away, tell me: what are the 3 most likely ways this goes wrong, what is the realistic timeline, what hidden costs am I not seeing, and what would I tell a friend who described this to me at 2 AM.
Explores consequences beyond the obvious first level.
I am considering this action: [describe]. Think beyond the first consequence. What happens after the first consequence? And after that? Map 3 levels of consequences. Most people stop at level 1  show me levels 2 and 3.
Builds the strongest version of the opposing view.
I disagree with [opposing view]. Instead of attacking the weakest version of their argument, construct the strongest possible version (steel man). Make their argument so good that someone who agrees with them would say 'yes, exactly.' Then respond to THAT version.
Identifies perspective-based blind spots.
Based on my reasoning about [topic], identify my blind spots. What am I not considering because of: my profession, my culture, my recent experiences, my social circle, my economic position? What would someone from a completely different background say about my reasoning?
Asks the questions you are avoiding.
I am dealing with [situation]. Ask me 5 questions that I do not want to answer but need to. The questions that make me squirm. The ones where my honest answer would change my plan. Do not let me deflect  just ask.
Independent judgment, not diplomatic hedging.
I have presented my case for [topic]. Give me your independent verdict. Not 'it depends'  a real assessment. Do I have a strong case? What percentage of my argument holds up under scrutiny? What is the weakest link? What would a neutral judge decide?
Chatgpt Prompt To Not Agree With Everything
ChatGPT and Claude
Well-structured written content in your chosen format, tone, and length. Includes headlines, body copy, CTAs, and optionally meta descriptions or SEO keywords.
Be specific about your audience, tone, and goal. Include a reference example if possible. Iterate by giving targeted feedback rather than starting over.
You can customize the prompt by adjusting the tone (professional, casual, authoritative), output length, format (bullet points, paragraphs, tables), and adding domain-specific context. Include examples of work you admire, brand guidelines, or specific constraints. For advanced use, chain this prompt with others for multi-step workflows.
This prompt helps you generate Chatgpt Prompt To Not Agree With Everything using AI. It provides a structured template you can fill with your specific requirements.
Absolutely. The prompt serves as a flexible starting point. Adjust the tone, length, format, and specific instructions to match your project.
Yes, all prompts on AI Bloater are completely free to use. You can copy, customize, and use them with any AI tool of your choice. No attribution or registration required.