Using AI · Growth Library
Use AI to Widen the Idea Set Before You Choose
Ask for meaningfully different directions, then judge them yourself
Try this
State the problem and constraints. Ask for several directions that must differ from one another in a named way. Cluster similar answers, discard obvious filler, and evaluate the remaining options against real constraints and evidence.
Check-in: Did AI add genuinely different options, or merely rephrase the same idea several times?
Evidence: Reviewed · source · Lee & Chung (2024) — An empirical investigation of the impact of ChatGPT on creativity
Review note: Reviewed against Lee & Chung (Nature Human Behaviour, 2024). Multi-experiment evidence supports higher judged creativity on several ideation tasks. The protocol is limited to option generation; feasibility, selection and implementation remain separate judgments.
AI brainstorming is most useful before convergence. Its job here is to increase the number of plausible paths you can inspect, not to crown a winner after reading half your situation.
1. Define the problem and constraints
Give the model the problem, what cannot change, and what success would look like. Without constraints, variety can become decorative nonsense.
2. Force meaningful variation
Request alternatives that differ by mechanism or trade-off: cheaper versus faster, manual versus automated, conservative versus aggressive, individual versus team, or another dimension that matters.
3. Cluster before judging
Group rephrasings of the same concept together. Keep only distinct approaches. This prevents a list of fifteen bullets from masquerading as fifteen ideas.
4. Apply human constraints
Evaluate the surviving options against cost, risk, evidence, ethics, dependencies, and your real objective. Generate broadly; decide narrowly.
Questions to consider
Why not ask AI for the best idea?
Because selection depends on context the model may not have. It is often more useful as an option generator than as the authority that chooses the option.
How do I get less repetitive ideas?
Force variation: ask for different mechanisms, target users, constraints, price levels, failure assumptions, or implementation paths. Then remove near-duplicates before judging.
Sources
Evidence state: practical. Sources are provided so the underlying material can be inspected directly.
Maintenance record
Review history
Current status: Reviewed. An editorial or evidence review is recorded and no later event changes that conclusion.
2026-09-15 · reviewed · Brali editorial agent Source: review-registry-using-ai-starters.json
Publish as a bounded divergence protocol: empirical evidence supports creativity gains in specific ideation tasks, while selection remains tied to the user's real constraints.
Browse the review ledger · Challenge or update this hack
Implementation observations can trigger a review, but they cannot change an evidence conclusion by themselves. Commercial relationships do not control review status or outcomes.
Continue from here
Choose the next useful path.
Article versions
Brali keeps substantive article history visible. Later reviews may refresh wording, sources, examples, or boundaries without silently replacing the record.
- Version 2026-09-15: first recorded publication in the migrated Brali corpus.
- Evidence review 2026-09-15: reviewed by Brali editorial agent; evidence state is
reviewed.
Canonical source record · Evidence state: reviewed.