Developers have been approving their own models as high-risk-cleared while a proof of concept becomes a governed production system. What is wrong with this practice?
Select an answer to reveal the explanation.
Short Explanation
If you're building it, you shouldn't be the one signing off on its risk. Independent review exists so builders don't certify their own work, and that's exactly what structured lifecycle review is meant to prevent.
Full Explanation
The right choice is the one saying that this practice removes the independent check structured lifecycle review is meant to provide, because the people building a system should not also be the ones certifying its risk. Independent review exists because builders are naturally invested in their own work progressing; a structured workflow separates the person accountable for building from the person accountable for approving. The choice saying the practice is acceptable if developers document their reasoning afterward is wrong because after-the-fact documentation does not substitute for an independent reviewer's judgement before the decision is made. The choice saying it only matters for models classified as low risk is wrong because self-approval is a problem at any risk level, and is most dangerous precisely where risk is highest. The choice saying it speeds up delivery, so the trade-off against independent review is generally worthwhile, is wrong because trading independent oversight for speed is exactly the trade-off structured governance is designed to prevent. For the exam, keep approval authority separate from the team that built the system.