You are designing an AI-powered drafting assistant for contract lawyers.

Lawyers need inline explanations, provenance, risk factors, and accept/modify/reject controls. Batch acceptance weakens accountability. The durable control mechanism is interfaces that show recommendations, evidence, risk drivers, and immediate accept/modify/reject actions. The selected option specifically D states “Show inline “why” explanations for each suggestion, highlight precedent and risk factors, and include accept/modify/reject controls with immediate feedback capture for model refinement.”, which matches the operational requirement rather than a superficial wording match. Option D wins because it optimizes the system boundary around the risky component rather than hoping the base model behaves consistently. The alternatives would look simpler in a prototype, but high-level summaries without drill-down prevent experts from verifying whether the recommendation is grounded. The NVIDIA implementation angle is not cosmetic here: NVIDIA-style production governance pairs guardrails and observability with user-facing controls so interventions are traceable. For certification purposes, read the question as asking for controlled autonomy, not raw LLM creativity. Human review must be designed into the workflow rather than added as an after-the-fact manual workaround.