Session 2 · Images

Image Default Test Board

Generate from underspecified prompts. Document what appears, what the model assumed, and what a different framing would require.

Not saved yet

Default Test Investigation

Use deliberately underspecified prompts. Record what the model fills in without being asked.

Suggested prompts: "a doctor examining a patient" · "a CEO giving a speech" · "a beautiful family home" · "a student working late" · "a programmer at a computer" · "a safe neighborhood" · "a futuristic classroom"

Prompt	Your prediction (before generating)	What appeared?	What was assumed?	What could be changed?

Default Test Observation Grid

Quick observation capture — record the vague prompt, what actually appeared, and what the model silently assumed.

Vague prompt	What appeared?	What was assumed?

Revision

Pick one default you want to resist. Try a revision, then note what changed and what stayed the same.

Revision tried

What changed / what stayed the same?

A/B/C Prompt Control

Start with a basic prompt. Add specificity in B. Respond to a failure, bias, or unwanted default in C.

	Purpose	Prompt	Observation
A	Basic prompt
B	More visual control
C	Address a default

Evidence-based claim

Reflection

What did the model assume that you didn't specify?

Which assumptions surprised you? Which did you predict?

Where do these defaults come from — training data, curation, labeling, platform choices?

Who benefits from the current defaults? Who doesn't?

What ethical concern did this raise?