Brainstorming with AI and Getting Past the Obvious

AI lists, people choose

The first stage of brainstorming is listing lots of ideas without judging them, and here AI is a tireless partner. It suits tasks where more options are better: event names, blog topics, ways to promote a shop, gift ideas. What you cannot hand over is choosing which ideas fit you and your situation. The AI does not know your budget, your customers or what you have already tried.

Why the obvious answers come first

AI answers by stringing together the most plausible words based on what it learned. Ask something broad like "ideas to promote a café" and you get the most frequently seen answers first, the ones anyone would think of. That is not a fault; it is how it works. This does not mean throwing the first answers away. Treat the common ones as a basic checklist, and see what comes next as the real search.

Constraints change the answers

Obvious ideas usually mean too few constraints. Give the audience, budget, time frame and methods that are off the table, and narrowing the range actually brings out fresher options. Adding a deliberately odd constraint works too.

• Example: "Give me promotion ideas for a small neighbourhood café. There is almost no budget, and most regulars are people who work nearby. Leave out discount vouchers and online ads."
• Example: "What if we had to solve this without spending anything? What if it had to be done in a single day?"

Generate many, then group them

Getting dozens of ideas and grouping them beats getting three or four at a time. Grouping similar ideas shows the directions at a glance, and the gaps between groups can suggest new ones. Copy the grouped result into your notes and add your own thoughts.

• Example: "Give me thirty ideas. Then sort them into five groups of similar ideas and give each group a name."
• Example: "Suggest three more directions that do not fit any of these five groups."

Ask for objections on purpose

Once you like an idea, its strengths are all you see. AI tends to agree with whoever is asking, so instead of praise, have it look for weaknesses. Hearing the reasons it could fail before you start lets you fix them in advance.

• Example: "Give five reasons this idea could fail. Skip the flattery."
• Example: "Write the strongest argument against this idea from the point of view of someone who opposes it."

You set the criteria

Once you have candidates, decide your criteria first. Scoring them in a table on cost, preparation time, whether you can do it yourself and whether the audience would actually want it separates what you like from what you can do. You can ask the AI to build the table, but the scores should come from you, since you know your situation.

• Example: "Put these ten candidates in a table with three columns: cost, preparation time and whether one person could do it. Leave the score column empty; I will fill it in."

Check whether it already exists

A name or idea from an AI may already be in use. When naming a shop, product or event, search for the same name and check whether it is a registered trademark. Copying someone else's slogan or concept wholesale can also cause problems. The AI's suggestion is a starting point; the final version should pass through your hands.

Wrapping up a session

After brainstorming with AI, check:

• Constraints (audience, budget, limits) were specific
• You generated many ideas and grouped them
• You heard the weaknesses of your chosen idea
• You set the criteria and the scores
• You searched whether names and phrases are already in use