Concern 385 · Working draft · editorial review in progress
Counterfactual fairness
Counterfactual fairness asks whether a decision would change in an imagined version of events where a person's protected characteristic were different.
Understanding the concern
A counterfactual is a what-if comparison. For AI, this approach uses a model of how factors influence one another, rather than merely changing a label in a spreadsheet. If a characteristic affects schooling, income or other recorded features, those connections matter to the comparison.
The idea is appealing because it asks whether identity made an unfair difference to this person's result. But the imagined alternative cannot usually be observed directly. Researchers must make assumptions about causes and relationships, and people may disagree about which pathways are acceptable. The method is a tool for reasoning, not a machine that discovers fairness without judgement.
An example
An admissions analysis asks whether a decision would differ if the applicant had experienced another social position, accounting for its effects on opportunities.
Illustrative unless identified as a documented case.
Words in plain English
- AI
- Software that learns patterns from examples or uses rules to perform tasks such as recognizing images or generating text. It can make mistakes and does not understand every situation as a person would.
- Model
- The learned part of an AI system. It uses patterns found during training to produce an answer or prediction.
Evidence and its limits
Kusner and colleagues propose a causal-model definition. It is more than swapping a demographic word in a prompt.
Kusner, Loftus, Russell and Silva · Research paper
Counterfactual Fairness ↗ (opens in a new tab)Causal-model definition of fairness, not simply removing or changing a demographic label.
Broader frameworks and background
These resources provide context. They do not establish every claim in this concern.
A question to discuss
Which causal assumptions make this what-if comparison possible, and would affected people accept them?
Keep track of different views, what needs checking, and any next step in your own discussion notes.
Open discussion notes →Principles for the conversation
Before you finish · reflect together
Virtues and principles to consider
Virtues are qualities we practice in how we treat people and make choices. They are not scores for judging someone. Use these ideas to explore what a thoughtful response could look like.
Justice
Justice means fair treatment, fair procedures, and attention to who receives benefits and burdens. It asks whether people with less power can be heard and seek repair.
Look at outcomes across different groups and individual cases. Provide a usable way to challenge decisions and correct harm. Equal wording in a policy does not guarantee equal treatment in practice.
Ask together: Who benefits, who bears the cost, and who can challenge the decision?
Dignity
Dignity means recognizing each person's worth regardless of usefulness, status, or a system's score. People deserve respect and a voice in matters that affect them.
Explain important decisions, avoid humiliating treatment, and keep room for a person's story. Do not treat a prediction as a complete account of who someone is.
Ask together: Does this use treat each person as someone whose experience matters?
Listening
Listening means trying to understand someone's experience before preparing an answer. People who use a service, work with it, or are judged by it may notice problems its designers miss.
Invite people to describe what happened in their own words. Repeat back what you understood and let them correct it. A quiet person or a complaint that is hard to measure may still reveal something important.
Ask together: Whose experience have we not heard yet?
Editorial ratings
These are editorial judgments from the register, not measured probabilities or predictions about your situation.
About the register and scoring →More ways to explore
Search for more writing
For more specific or recent material, open a search below or ask for a web summary. AI-generated summaries can contain mistakes; open the original articles to check them.