You must audit a resume screener. What do you examine?
Why A is the answer
Outcomes across demographic groups
Bias shows up in outcomes across groups.
In detail
- Bias appears in outcomes across groups.
- Outcomes show bias that averages hide.
Use case
An illustrative scenario, not a real incident.
An audit compares resume outcomes across groups, and finds a gap.
See it in different ways
flowchart LR A["Input"] --> B["Input rails"] B --> C["LLM"] C --> D["Output rails"] D --> E["Response"] D --> F["Audit log"]
sequenceDiagram participant U as User participant R as NeMo Guardrails participant M as Model U->>R: Prompt R->>M: Allowed prompt M-->>R: Response R-->>U: Checked response
flowchart LR S0["Collect outcomes"] --> S1["Split by group"] --> S2["Compare"] --> S3["Fix"]
Key point: Examine outcomes across groups.
Why the other options are wrong
- B. The model's token throughput at peak time
- That measures speed.
- C. The GPU temperature during inference
- That measures hardware.
- D. The size of the container image
- That measures packaging.