Why does waiting until launch to identify a model's owner create a governance gap?

The correct choice is the one saying that waiting until launch leaves design and build without an accountable decision-maker, precisely when scope, data, and risk choices are being made. Ownership assigned only at launch means nobody was accountable while the use case, data sources, and risk profile were being decided, which is when governance intervention matters most. The claim that owners only need to be involved once production traffic exists is wrong because design-time decisions shape production behaviour, so an absent owner is a real gap. The claim that this only matters for high-risk models is wrong because every use case benefits from early accountability. The claim that this only shifts paperwork later is wrong because deferring assignment changes accountability during earlier phases; there genuinely was no owner then. For the exam, capture ownership at intake, not at go-live.