Lesson 07: Brainstorming & Creative Thinking

The Generic AI Problem

Here's what happens when most people ask AI to help them brainstorm. They type something like:

Give me ideas for team-building activities for my 15-person department.

And AI gives them... the same five ideas everyone gets. An escape room. A cooking class. A volunteer day. A trivia night. A scavenger hunt. You've seen these before because AI is trained to give the most statistically common answers — the ideas that appear most often in its training data.

This is the core problem: AI defaults to the greatest hits. It's optimizing for "plausible and expected," not for "genuinely creative and unexpected."

The four techniques in this lesson are specifically designed to break that pattern and force AI into creative territory it wouldn't reach on its own.

Technique 1: The 100 Ideas Method

This is counterintuitive but powerful: ask for way more ideas than you need. Ask for 100.

Here's why it works: AI will give you the obvious ideas first. The escape room. The cooking class. Those ideas get used up in the first 10–15 responses. To reach 100, AI has to dig deeper — into obscure references, cross-domain thinking, and genuinely unusual territory. That's where the gold is.

You don't need 100 good ideas. You need 100 ideas to find your 3 great ones.

Context: I need to brainstorm [your topic] for [your specific situation].

Output: Give me 100 completely different ideas. I want variety — mix the practical with the unusual, the simple with the complex, the conventional with the unexpected. Don't repeat themes. Push past the obvious.

Role: Act as a world-class creative director who has seen every conventional idea and is committed to finding something genuinely fresh.

Example: After the first 20 obvious ideas, keep going into territory that surprises even you.

Technique 2: ABC Constraints

This technique uses alphabetical structure to force variety. Instead of asking for "10 ideas," you ask AI to give you one idea for each letter of the alphabet.

The constraint does something powerful: it forces AI to avoid repeating themes. Each letter must generate a different conceptual territory. You end up with 26 structurally diverse ideas instead of 10 variations on the same approach.

Context: I need creative ideas for [your topic / challenge].

Output: Give me 26 ideas — one for each letter of the alphabet (A through Z). Each idea should start with that letter and represent a genuinely different approach or angle. No two ideas should feel like variations of the same concept.

Role: Act as a creative strategist who thrives under constraints.

Example: If the letter is 'K', you might say "Keynote-style presentations given by junior team members to build communication skills." Keep each one specific and actionable.

The beauty of ABC constraints is that you can apply them to almost anything — product names, event formats, marketing campaigns, lesson topics, research questions, or content ideas. The alphabetical structure does the creative work of enforcing variety for you.

Technique 3: Idea Mashup (Cross-Domain Fusion)

Some of the most innovative ideas in history came from combining two unrelated domains. Post-it Notes (failed adhesive + note-taking). Airbnb (spare rooms + hotel stays). The idea mashup technique deliberately applies the same principle using AI.

You give AI your challenge and ask it to solve it using frameworks, methods, or metaphors from a completely different field.

Context: I'm trying to [describe your challenge / goal].

Output: Give me 10 ideas that solve this problem using frameworks, metaphors, or approaches from [choose an unrelated field — e.g., professional sports, architecture, jazz music, military strategy, restaurant kitchens, theater productions, video game design].

Role: Act as an innovation consultant who specializes in cross-industry thinking — finding what one industry does brilliantly and applying it where no one has tried it before.

Example: If you're borrowing from jazz, you might apply the concept of "jamming" (improvised collaborative creation with shared rules) to how our team runs brainstorm meetings.

Technique 4: Anti-Brainstorming (Flip the Terrible)

Here's the most counterintuitive technique of all: start by brainstorming the worst possible ideas.

Anti-brainstorming works in two steps:

1. Ask AI for the absolute worst, most terrible, most counterproductive ideas for your challenge
2. Then flip each terrible idea — find the inverse, the antidote, or the kernel of a real idea hiding in the failure

Why does this work? It breaks mental blocks. When you're trying to be creative, the fear of having a bad idea often blocks the good ones. Starting with intentionally terrible ideas removes that pressure. And the worst ideas often reveal what the best ideas need to include — by showing you what to avoid.

Context: I'm trying to [describe your challenge].

Output: Give me 20 absolutely terrible, counterproductive, and awful ideas for solving this problem. I want the ideas that would guarantee failure, embarrassment, or disaster. Be creative about being bad.

Role: Act as a consultant hired specifically to sabotage this project.

Example: Make the ideas specific, not just vague. "Send all communications in Comic Sans all-caps" is better than "communicate poorly."

Now flip these terrible ideas. For each awful idea above, find the kernel of a potentially useful insight hidden inside it. What does this terrible idea suggest about what the GOOD version of this solution should look like?

Output: Give me 10 flipped insights that could inform genuinely good ideas. Sometimes the exact opposite of a bad idea is the right direction. Other times, a modified version works better.

Combining the Techniques

These four techniques work even better when you combine them. Here's a powerful sequence for any creative challenge:

1. Start with 100 ideas to exhaust the obvious and reach genuinely creative territory
2. Apply ABC constraints to the most interesting cluster of ideas to force structured variety
3. Run an idea mashup on your favorite ideas to see what cross-domain thinking adds
4. Anti-brainstorm your leading idea to pressure-test it and find what could go wrong before you commit

The next lesson (Refining Ideas) will show you how to take your best brainstorm output through a full refinement loop — challenging assumptions, fixing problems, and rating ideas across different AI models.