AI AI Tools Guidebook
AI Tool Tutorials

ChatGPT for Brainstorming

Free up your thinking by externalizing it — but only if you push back.

Published: Author: AI Productivity Guide Team 🌐 查看中文版本

Brainstorming alone tends to converge on the first three obvious ideas. Brainstorming with ChatGPT, done badly, does the same — except now you also believe a machine validated them. This guide is for the founders, writers, and operators who want to use ChatGPT as a sparring partner that pushes you past the obvious, not as a yes-man that polishes your first instinct.

What this covers

Use ChatGPT to externalize your thinking, generate radically different angles, and stress-test them — but only if you push back on every answer. The pattern is: diverge wide, pick three, attack them, decide.

Key tools and concepts:

  • ChatGPT: OpenAI’s conversational AI assistant — the product that brought the GPT models to a mass audience.
  • Divergent vs convergent prompting: Divergent asks for many wildly different angles; convergent narrows down. Beginners mix them in one prompt and get mush.

Who this is for

Founders evaluating product directions, writers stuck on framings, PMs naming features, operators comparing trade-offs. Anyone making a decision where “what are the options I am not seeing?” is the bottleneck.

When to reach for it

Naming a product, framing a launch, picking between three structural directions, mapping risks, or rebuilding a stuck argument. Skip it when the decision is purely about facts — that is research, not brainstorming.

Before you start

  • Write the problem in one sentence. If you cannot, brainstorming is premature — you are still framing the problem.
  • Note your own first three ideas before opening ChatGPT. Otherwise its suggestions anchor you and you stop generating your own.
  • Decide what “good” looks like. “Something I have not thought of” is a fine criterion; “the best idea” is not.

Step by step

  1. State the problem in one sentence with the constraint: “I am [role] choosing [decision] given [constraint]. List 10 wildly different angles, no overlap.”
  2. Read all 10. Cross off the ones that overlap with your pre-written first three. The valuable ones are usually 4-10.
  3. Pick the 3 most interesting (or most uncomfortable). Ask the model to elaborate each into a 5-line pitch with one risk.
  4. Attack each pitch. “What is the strongest reason this is wrong? Who would this fail for?”
  5. Force a synthesis: “Combine the best part of pitch A with the risk-mitigation of pitch C into a new option D.”
  6. Stop when you have a top-2 you can defend in writing. Save the chat — the rejected ideas often resurface later.

Example prompt

I am a solo founder choosing the wedge for my new ops-tools
product. Constraint: I have 8 weeks of runway to validate.

Give me 10 wildly different wedges — each on a different axis
(audience, distribution, format, business model, integration).
No overlap. Skip the obvious 3 (Notion-style template, Slack
bot, Chrome extension). For each: one sentence pitch, target
user, and the first thing you would build in week 1.

First-run exercise

  1. Pick a real decision you have been stuck on for at least a week. Stuck means “I keep thinking about it without converging.”
  2. Run the diverge prompt. Note which of the 10 angles you reflexively dismiss — those are often the ones worth a second look.
  3. Pick the 3 most interesting (not the 3 safest). Run the elaborate-and-attack loop.
  4. Write your decision in one paragraph, including which angle you chose and what would change your mind.

Quality check

  • Did the model give you angles you would not have generated alone? If 9 of 10 feel obvious, your problem statement was too narrow.
  • Are the attacks on each pitch substantive (specific user, specific failure mode) or vague (“might not work”)?
  • Could you defend your final choice to a skeptical investor or editor without re-running the chat?

How to reuse this workflow

  • Keep a brainstorm-decisions.md log: problem, 10 angles, chosen path, what would change your mind. Re-read quarterly.
  • Save the diverge prompt above as a template. Replace the role, decision, constraint, and dismissed-three each time.
  • For team brainstorms, pre-run this solo, bring the 10 angles to the meeting, and skip the warm-up phase.

One-sentence problem with constraint → 10 angles (no overlap) → pick 3 → elaborate with risk → attack each → synthesize a 4th option → decide and write it down. The whole loop fits in 25-40 minutes for medium-stakes decisions.

Common mistakes

  • Stopping at the first 3 ideas — those are the ones you would have generated solo anyway.
  • Not pushing back. The model will agree with whatever you escalate. Disagree on purpose to find the weak spots.
  • Asking generic prompts like “ideas about marketing.” Constraints (audience, budget, timeline) are what produce specific ideas.
  • Letting ChatGPT pick which angle is best. That is your job — the model has no skin in your decision.
  • Using brainstorming when the bottleneck is actually research. If you keep saying “I do not know enough,” go find data, then brainstorm.
  • Throwing away the rejected ideas. Save the chat; 20% of “obviously bad” ideas look great six months later.

FAQ

  • Should I tell the model what I have already considered?: Yes — list your dismissed-three so it skips them and generates genuinely new ones.
  • Reasoning model or fast model?: Fast model for the diverge step (variety matters more than depth), reasoning model for the attack step.
  • What if all 10 angles feel weak?: Your constraint is probably wrong. Loosen one constraint and re-run, or attack the problem statement itself.
  • Is brainstorming with AI worse than with humans?: Different. AI is faster and gives more variety; humans bring lived constraints and emotional stakes. Use both, not one.

Tags: #ChatGPT #Tutorial

Related Articles