Module 12 · Tier 5 — Workflow Integration

AI for Creativity & Ideation

Brainstorming, concept development, and using AI as a thinking partner.

10 min Tier 5 — Workflow Integration

Why This Matters

Creative blocks are expensive. Waiting for inspiration is slow. AI changes the economics of ideation: it eliminates blank-page paralysis and produces a volume of starting points that manual brainstorming rarely matches. The constraint is that AI generates plausible-sounding ideas more reliably than genuinely novel ones. This module is about getting the volume and using your judgment to find the signal.

AI is excellent at: generating large volumes of options quickly, combining concepts in ways that surface non-obvious connections, exploring variations on a theme, thinking from a different perspective than your default, and getting past blank-page paralysis.

AI is not excellent at: generating truly original ideas (it recombines what exists), judging which ideas are actually good, and providing the domain judgment to know what's feasible.

The creative workflow where AI adds the most value: you bring domain knowledge and quality judgment; AI brings volume and variation.

Full access is for AIQ members

Unlock all 56 lessons, the certificate pathway, and the SociA|~ community.

56 lessons across all three course tracks
AIQ certification on completion
SociA|~ Society community access

Unlock full access →

The Concept

What AI Does and Doesn't Do for Creativity

AI is not excellent at: generating truly original ideas (it recombines what exists), judging which ideas are actually good, and providing the domain judgment to know what's feasible.

The creative workflow where AI adds the most value: you bring domain knowledge and quality judgment; AI brings volume and variation.

The Diverge-Converge Pattern

The most effective AI-assisted creative workflow has two phases:

Diverge: Generate as many options as possible without judgment. Ask AI for 20 options, not 5. Ask for ideas from a different perspective, a different industry, a different constraint. Remove the word "good" from your prompt — just generate.

Converge: Apply your judgment to the full set. Which ideas are genuinely interesting? Which are plausible-but-boring? Which combine to make something better? Human judgment drives convergence. AI can help by steelmanning your selected ideas or identifying weaknesses.

Ideation Prompt Patterns

Volume generation: "Give me 20 [ideas/angles/approaches] for [problem/task]. Don't filter for quality — I want quantity."

Perspective shift: "How would [a different industry/a 5-year-old/a skeptic/someone in 1990] approach this problem?"

Constraint injection: "Generate solutions to [problem] with this constraint: [unusual constraint]. The constraint is non-negotiable."

Combination: "Here are two ideas I like: [A] and [B]. Generate 10 variations that combine elements of both."

What if: "What if [assumption about this problem] were false? What would change about the approach?"

Keeping the Human in the Loop

The failure mode: accepting AI's first interesting-sounding idea without applying domain judgment. AI generates ideas that sound creative and are actually generic — they just use unusual vocabulary to describe common approaches.

The discipline: evaluate every AI-generated idea against your domain expertise. "Is this actually novel in my field? Has this been tried? What would make it actually work?"

Volume over curation

A product team is naming a new feature. Manual brainstorm: 8-12 names in 30 minutes, heavily influenced by what the first person says.

AI-assisted: "Give me 40 possible names for a feature that [description]. Include: 5 that are purely descriptive, 5 that are metaphorical, 5 that reference the benefit, 5 that are short enough to be a keyboard shortcut, and 15 in whatever style feels most interesting to you."

Result: 40 names in 90 seconds. Most are forgettable. Three are interesting. One combination of two names becomes the actual choice.

The creative breakthrough came from volume that made an unexpected combination visible — not from AI being creative, but from AI generating enough that human pattern recognition could find something good.

Hands-On Exercise

Run a diverge-converge session on a real creative problem

ClaudeChatGPT

Choose a real creative challenge you're working on — a name, a concept, a problem to solve, a campaign idea. Diverge: 1. Ask AI for 25 ideas (no quality filter) 2. Ask for 10 more from a completely different angle or industry 3. Ask for 10 that violate your first assumption about the problem Converge: 4. Without re-reading AI's list, write your own 5 favorite ideas 5. Now read the AI list: what surprised you? What did you miss? 6. Pick 3 AI ideas and 2 of your own and ask AI to combine them Final: what came out of the session that you wouldn't have reached without it?

The goal of diverge is volume, not quality. Suspend judgment until converge. This is the discipline that makes the process work.

Active Recall

Before moving on — close this lesson and answer these from memory. Then come back and check. Testing yourself (not re-reading) is how this sticks.

1 What does AI do well in creative work, and what does it do poorly? How does understanding this change how you use it for ideation?

2 Describe the diverge-converge pattern. What is the risk if you skip the diverge phase and ask for 'the best 3 ideas' immediately?

Reflection

Where in your work do you most often feel creatively stuck or limited by blank-page paralysis? How would you apply the diverge-converge pattern to that specific situation?

Key Takeaway

AI adds creative value through volume and variation, not novelty. Separate diverge (quantity, no judgment) from converge (quality, your judgment). Ask for more options than you think you need. The valuable idea is often in the combination, not the original generation.