
Model Dignity Check
From Chapter 9 of AI and the Art of Being Human
What it is
A pre-launch checklist for ensuring AI systems preserve human dignity and catch blind spots before they scale.

Where you would use it
Before any AI system or feature goes live
When reviewing existing AI implementations
During design phases, to catch problems early
When updating or retraining AI models



The Tool
Five Questions to ask before launch
Who becomes invisible when we optimize? Name specific people, not categories—“elderly residents in walk-ups” not “some users”
What “normal” is baked into the training data? Every dataset tells a story about who matters—who’s reality shaped this system?
How does this perform for our most vulnerable users? Test on edge cases—the users with least power, resources, or technical literacy
Can affected humans understand and contest decisions? Opacity breeds distrust—is there a clear path to challenge the algorithm?
Does this strengthen or erode human agency? Are we augmenting human judgment or replacing it?

How to use it
Before launch, document written answers to all five questions. Be specific—vague answers hide real problems. If any answer troubles you, redesign before deploying. This isn’t a compliance checkbox but a discipline for catching what pure optimization misses. Run this check again whenever you update or retrain the system.
See it in practice: Who Becomes Invisible When We Ship →
FAQ
Add paragraph text. Click “Edit Text” to update the font, size and more. To change and reuse text themes, go to Site Styles.
The Model Dignity Check is a prompt for considering whether an AI interaction or output treats people with respect. It helps you look beyond whether a result is technically useful and consider its effect on human dignity.
Use it before sharing AI-generated content about people, deploying an AI-supported process, or relying on a model's description or assessment of someone. It is also useful when an output feels demeaning, reductive, or unfair.
Consider whether the output respects people's agency, context, and individuality. Look for stereotypes, unsupported assumptions, demeaning language, or conclusions that reduce a person to a label. Revise or reject outputs that fail to meet the standard of respect you expect.





