liveBook · Manning

Fairness metric

As explained in AI Governance

A fairness metric is a quantitative measure of disparities or equitable outcomes across people or groups.

A fairness metric measures disparities or equitable outcomes across people or groups. Fairness is not captured by a single metric: the relevant lens depends on the use case. section 6.8

Use fairness metrics as launch criteria, then monitor them actively after deployment. Passing security and privacy checks does not establish that a system is fair. section 6.8

How AI Governance covers it

How deep it goes
Defines fairness as use-case-dependent and sets metrics as launch criteria, without specifying particular measures.
What you can do after
Evaluate launch readiness and monitor fairness after deployment using metrics suited to the use case.
How it connects
Connects fairness metrics to GenAI system launch decisions and ongoing monitoring.

Also explained in

Compare how 2 books explain fairness metric →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of AI Governance. The section links point to the parts of the book it draws on.