chatgpt-brainstorming
ITIAN ChatGPT Academy
Technology Simplified — Solutions That Work
Brainstorming and Developing Ideas with ChatGPT
Generate varied possibilities, escape the first obvious answer, evaluate ideas against useful criteria and turn a promising direction into a small, testable next step.
Learning Outcomes
By the end of this lesson, you should be able to:
Frame a Useful Challenge
Describe the goal, audience, constraints and boundaries without prescribing the solution.
Generate Real Variety
Use several creative lenses instead of accepting a list of near-duplicate ideas.
Evaluate Transparently
Shortlist ideas with named criteria, assumptions and evidence rather than enthusiasm alone.
Move to a Safe Experiment
Turn the strongest option into a small, reversible test that produces useful learning.
Separate imagination from judgement
During idea generation, premature criticism narrows the field. During selection, uncritical enthusiasm hides cost and risk. Use two distinct modes: first diverge to create variety, then converge to compare and choose.
ChatGPT supplies possibilities. You supply context, values, lived experience and final judgement.
The ITIAN Idea-to-Action Cycle
A strong brainstorm continues beyond the first list.
Frame
Define outcome, audience and constraints.
Expand
Generate many options without ranking.
Stretch
Use new lenses to force greater variety.
Cluster
Group themes and remove duplicates.
Select
Score a shortlist against criteria.
Test
Choose one small experiment and learn.
Eight Lenses for Less-Obvious Ideas
Ask for a separate round through each lens rather than one enormous list.
Audience Lens
How would the answer change for beginners, experts, families or people with access needs?
Constraint Lens
What becomes possible with one hour, no budget, a small team or existing tools only?
Reverse Lens
What would make the problem worse—and what opposite principles does that reveal?
Analogy Lens
How do libraries, museums, gardens, games or other fields solve a similar problem?
Combination Lens
Which two partial ideas could be combined into a stronger option?
Scale Lens
What is the smallest useful version, and what would a ten-times-larger version require?
Channel Lens
How could the experience work in person, online, by email, video, audio or print?
Risk Lens
Which option is safest, most reversible, easiest to test or least likely to exclude people?
Interactive Idea-Sprint Builder
Create a structured brainstorming prompt locally. Use non-sensitive information and review the prompt before copying it.
Example: Improve Course Discovery
Opening Prompt
Help me brainstorm ways for adults returning to learning to find a suitable ITIAN short course within three minutes. Generate 15 ideas without ranking them. Include practical, accessible and unusual options. Use the audience, constraint, analogy and channel lenses. Do not invent visitor research—label assumptions clearly.
Useful Follow-Up Rounds
- Stretch: “The first list is too similar. Add ten ideas based on libraries, streaming services and guided interviews.”
- Cluster: “Group all ideas by underlying approach, not wording.”
- Challenge: “For each cluster, identify who it may exclude and what evidence is missing.”
- Select: “Score the best six against learner value, accessibility, effort and reversibility.”
- Test: “Design a one-week test for the strongest option without rebuilding the website.”
From Long List to Responsible Shortlist
Use a matrix to make the reason for selection visible. Scores begin discussion; they do not replace judgement.
| Criterion | Question to ask | Evidence needed |
|---|---|---|
| Value | Does this solve the intended person’s real problem? | User feedback, observation or demand |
| Fit | Does it support the goal, brand and ethical boundaries? | Strategy and policy check |
| Accessibility | Who could be excluded, and can barriers be reduced? | Accessibility review and diverse testing |
| Feasibility | Are time, skills, tools and budget realistic? | Estimate from the people doing the work |
| Risk | What could fail, harm trust or create unintended effects? | Risk review and safeguards |
| Learning | Can a small test produce useful evidence quickly? | Clear hypothesis and success measure |
Ask ChatGPT to show its assumptions
Request separate columns for known facts, assumptions, questions and evidence needed. An attractive idea should not be presented as proven demand, technical feasibility or legal approval.
Similarity Is Not Variety
Twenty differently worded versions of one idea are still one direction. Ask for categories, opposing approaches and a deliberate second round.
Novelty Is Not Usefulness
An unusual idea can be exciting but impractical, inaccessible or irrelevant. Keep novelty separate from selection criteria.
Confidence Is Not Evidence
ChatGPT may invent statistics, trends, competitors or customer needs. Verify factual claims and conduct real research before investing.
AI Is Not the Whole Room
Include the people affected, subject experts and delivery team. Human perspectives reveal context and consequences a generated list may miss.
Recommended: 16:9 • Captioned • Evaluation matrix shown
Future Idea Workshop
This video will demonstrate a complete six-stage idea sprint.
- Frame a challenge without prescribing the answer.
- Use several lenses to improve variety.
- Cluster and evaluate transparently.
- Finish with one small experiment.
Practical Activity: Run a Complete Idea Sprint
Choose a real, low-risk challenge that does not require confidential information.
0 of 10 completed — begin by framing one challenge.
Knowledge Check
Answer all five questions, then check your result.
Official Sources and Further Reading
Reviewed against current first-party OpenAI guidance on 13 July 2026.
OpenAI guidance used in this lesson
- ChatGPT Capabilities Overview — creative suggestions, conversation context and available tools.
- Prompt engineering best practices — clarity, context and iterative refinement.
- How do I create a good prompt? — focused requests, multiple options and follow-up prompts.
- Does ChatGPT tell the truth? — limitations, critical assessment and verification.
Lesson Summary
Seven ideas to remember
- Frame the challenge, audience, desired outcome and constraints.
- Separate idea generation from idea selection.
- Use several creative lenses to produce genuine variety.
- Cluster related ideas and remove duplicates.
- Evaluate a shortlist using explicit criteria and real evidence.
- Include affected people and relevant experts in the decision.
- Finish with a small, reversible experiment and learn from it.