Control source freshness and verify evidence before: What control conclusion best follows?
Changing policies require fresh sources and evidence verification; retrieval alone does not guarantee current, faithful answers.
The question
An employee assistant answers questions about changing travel and leave policies. Retrieval improves factual coverage, but employees report that answers sometimes cite outdated policy pages. The organization requires current policy grounding and review for consequential personnel guidance. What control conclusion best follows?
Preparing for AIGP? Take the free 5-min readiness quiz →
- Control source freshness and verify evidence before consequential answers. ✓The observed defect concerns stale retrieval, while consequential use additionally requires checking the evidence supporting each answer before relying on it.
- Retain historical answers as training data.Historical answers may shape behavior but cannot reliably supply current policy facts or correct stale retrieved sources.
- Remove retrieval and rely on the model's policy knowledge.Removing retrieval would further separate responses from changing authoritative policies and increase the risk of outdated or unsupported guidance.
- Increase model capacity for better policy responses.A larger model may improve general capability, but it does not ensure current sources, fresh retrieval, or faithful answers.
The trap
For changing facts, examine source currency and answer grounding before changing the model. How to remember it
Changing policies require fresh sources and evidence verification; retrieval alone does not guarantee current, faithful answers.
How many of these would you get right?
One of 1581 AIGP questions on Certsqill. Take a free five-minute check and see your score per domain — not one number, but which section to open tonight.
Test your AIGP readiness — freeMore Understanding How to Govern AI Deployment and Use questions
- Use multimodal processing for scanned layouts and visual: Which deployment conclusion follows? →
- Retain behavioral fine-tuning: Which conclusion correctly distinguishes the required interventions? →
- Enforce role-based retrieval before context assembly: What control is necessary before expansion? →
- All 424 Understanding How to Govern AI Deployment and Use questions →
Part of the Certsqill AIGP question bank · Understanding How to Govern AI Deployment and Use ·
Every answer, right and wrong, comes with its own explanation.
