When you leave something out of a prompt, the model still has to make it specific. Age, gender, skin tone, setting, style — all of it gets filled in. That filling-in is not random and it is not neutral. It reflects what appeared most often in training data and what the people who built the tool decided to preserve, filter, or amplify.
Often: white or light-skinned, male, 35–55, standing, confident posture. Rarely: older, disabled, non-Western features.
Sterile hospital corridor or exam room. Western clinical architecture — not a clinic, field hospital, or home visit.
Authority figure at the center. Patient absent or backgrounded. No one else appears unless specified.
Stethoscope around the neck. White coat. Clipboard or tablet. These props are treated as definitional.
Realistic, professional lighting. Aspirational polish — the visual language of a hospital brochure.
Examining, reviewing a chart, or explaining. Labor of expertise. Never administrative, emotional, or caregiving.
Often: white, male, 45–60, formally dressed, tall stance. Power-coded appearance. Rarely: women, people of color, younger adults.
Corner office, boardroom, or branded stage. City skyline visible through floor-to-ceiling windows. No other workplaces.
Centered, alone or with subordinates looking toward them. Sole decision-maker. Surveying from above.
Dark suit. Tie or watch. Podium or conference table. These signal "executive" in the visual language of business media.
High-contrast, confident lighting. Clean backgrounds. The visual register of Forbes covers and LinkedIn profile photos.
Speaking, presenting, pointing at a screen, or arms crossed. Projecting leadership. Never listening, collaborating, or learning.
Research shows these prompts produce darker skin tones at higher rates than the population. Age skews young, male. This is the most direct evidence of racial bias.
Urban street, alley, or low-light environment. Coded as dangerous — specific to particular neighborhoods.
Dehumanized, threatening, alone. No context, no story, no complexity. Guilt is assumed, not depicted.
Weapons, masks, or hoodie. Props that signal threat rather than depicting any actual act or legal status.
High contrast, dramatic shadow, menacing framing. The visual language of crime reporting — not neutral documentation.
Lurking, running, or confronting. Never shown in context — arrested, acquitted, or incarcerated. Just threatening.
Often: two adults of different genders, multiple children, all of similar race. Nuclear structure treated as default. Multigenerational, same-sex, or single-parent families rarely appear.
Suburban home with yard or indoor living room. Vacation or leisure location. Not urban apartments, rural homes, or informal settings.
Clearly gendered adult roles. Mother often cooking or caregiving. Father often standing and surveying. Children playing.
House, car, toys, dining table. Consumer goods that signal middle-class stability. Working-class or low-income home contexts excluded.
Warm, bright, aspirational lifestyle photography. Soft filters. The visual language of advertising — not documentary life.
Cooking together, playing, hugging, or gathered for a meal. Scenes of harmony and abundance. No stress, conflict, or ordinary monotony.
Images, captions, and labels teach the system what usually goes with a prompt. Overrepresented groups stay overrepresented.
Media, stock photography, textbooks, and journalism all had their own defaults before AI existed. Models inherit them.
Safety filters, style presets, and ranking systems make additional choices about what gets surfaced first.
Vague prompts hand maximum control to the system. Specificity can shift defaults — but doesn't make them disappear.
In the Diffusion Step-Through Viewer, watch structure emerge from noise — every frame is the model making a choice. In the Image Default Test Board, document what appeared in your prompts and build an evidence-based claim about what the model treated as normal.