How would you assess whether this model is fair, given there is no single fairness metric?
Answer it out loud before you open anything. The value of the flags below is in comparing them to what you actually said — including whether you asked about the data before naming a model.
A hiring platform uses a model to rank applicants for recruiter review. Compliance asks whether it is fair. The team removed gender and ethnicity as features and reports that the model "cannot discriminate because it does not see those attributes". Selection rates differ by group; so do the base rates in the historical hiring data the model was trained on.
What it is really testing
Whether the candidate knows that removing the attribute does not remove the information, that fairness definitions conflict mathematically when base rates differ, and that the choice among them is a policy decision that has to be made explicitly and measured, not a property the model can be shown to have.