Responsible AI

Fairness & Bias

Bias can enter through data, labels, objectives, deployment context, and feedback loops. Fairness is not one universal metric; it is a product, policy, and measurement decision shaped by the harm model.
  • Historical data can encode historical unfairness
  • Labels can be biased proxies
  • Removing sensitive attributes is not enough
  • Fairness metrics can conflict
  • Slice evaluation is the minimum bar
  • Fairness requires governance, not just model tuning
Bias entry points
StageHow bias enters
Data collectionSome groups are underrepresented or measured differently
LabelsHuman/system labels encode prior decisions
FeaturesProxy variables reconstruct sensitive attributes
ObjectiveMetric rewards behavior that harms a group
DeploymentUsers interact with the system differently
Feedback loopModel decisions shape future data
Sources
  • Machine Learning Crash CourseML Fairness
  • Artificial Intelligence: A Modern ApproachAI ethics and risks
  • Made With MLResponsible AI