1
Weak motive problem: instead of asking for 'a deeper backstory,' the writer requests five different obligations that could force the protagonist to protect a rival and compares how each changes the midpoint choice.
8 AI Character Brainstorming Examples
Use these examples to study structure, specificity, tone, and variation. They are demonstrations—not claims about a real person or organization unless the example itself makes that explicit.
Before you copy
What to notice in the examples
A strong AI-assisted character brainstorming session produces multiple meaningfully different possibilities that can be tested against the story's conflict, theme, setting, existing canon, and character agency. The writer should be able to explain why the retained traits change behavior and plot rather than merely making a profile longer.
- Lock known canon first: role, age or life stage where relevant, relationships, setting constraints, established facts, and anything the AI must not alter.
- Brainstorm around causal dimensions such as motive, fear, obligation, contradiction, resource, relationship pressure, and difficult choice rather than decorative traits alone.
- Ask for several contrasting possibilities and force each option to show how it changes behavior under pressure.
- Separate generated possibilities from accepted canon so rejected ideas do not leak into later prompts as if they were facts.
- Run a human review for stereotype, cultural flattening, accidental imitation, continuity, plausibility, and whether the character still serves the writer's own story.
2
Predictable antagonist problem: AI proposes contrasting values the antagonist genuinely protects; the writer keeps the option that makes the conflict principled rather than merely cruel.
3
Flat friendship problem: the writer asks for three incompatible needs the friends could have in the same scene and tests which one produces both affection and friction.
4
Canon-safe fantasy example: the writer supplies the existing magic rules and asks for fears that arise from those rules; any suggestion requiring impossible magic is rejected.
5
Character contradiction example: a cautious medic wants institutional approval but repeatedly breaks minor rules to protect vulnerable patients; the contradiction predicts behavior instead of adding a random quirk.
6
Voice-preparation example: AI generates possible worldview tensions, but the writer develops actual dialogue independently so generated phrasing does not homogenize character voice.
7
Representation check example: a generated identity detail introduces a stereotype-dependent motivation; the writer removes it, researches the context independently, and rebuilds the conflict from story-specific pressures.
8
Continuity workflow: after selecting one motive, the writer updates the canon sheet and starts the next prompt with accepted facts only, preventing discarded alternatives from becoming accidental continuity errors.
Keep the underlying decision or pattern, then replace the subject, evidence, relationship, constraints, and tone with details that belong to your situation. If your final line still works after swapping only one noun, it may be too close to the example.