NCA-GENL Trustworthy AI Practice Question
A healthcare company is deploying an LLM to assist with clinical documentation. To ensure Trustworthy AI, they must implement measures to detect and mitigate hallucinations that could lead to incorrect medical records. Which two actions should they take? (Choose two.)
⚠ Common exam trap
The trap here is assuming that larger models or fine-tuning automatically reduce hallucinations, when in fact they can still generate confident errors without external verification.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Deploy a fact-checking module that cross-references generated statements against a trusted medical database.
To mitigate hallucinations in clinical documentation, grounding responses in verified guidelines via RAG and adding a fact-checking module against a trusted database are effective. These actions ensure outputs are based on authoritative sources and verified before use. Other options like larger models, fine-tuning, or higher temperature do not reliably reduce hallucinations and may introduce new risks.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Deploy a fact-checking module that cross-references generated statements against a trusted medical database.
Why this is correct
A fact-checking module acts as a post-generation verification layer, comparing the LLM's output against a trusted medical database to flag or correct hallucinations. This directly addresses the risk of incorrect medical records by catching errors before they are finalized. It complements other methods and ensures that only verified information is retained, enhancing trustworthiness.
- ✓
Implement a retrieval-augmented generation (RAG) system that sources from verified clinical guidelines.
Why this is correct
RAG grounds the LLM's responses in authoritative, up-to-date clinical guidelines, reducing the likelihood of hallucinations by providing factual context. In a healthcare setting, this ensures that generated documentation aligns with verified medical knowledge, directly mitigating the risk of incorrect records. It is a proactive measure that enhances trustworthiness by anchoring outputs to reliable sources.
- ✗
Set the model's temperature to a higher value to encourage more diverse outputs.
Why it's wrong here
Increasing temperature makes the model's outputs more random and creative, which typically increases the likelihood of hallucinations. In clinical documentation, where accuracy is paramount, higher temperature is counterproductive. Lower temperatures are generally preferred to promote deterministic, fact-based responses, so this action would worsen the problem rather than mitigate it.
- ✗
Use a larger LLM with more parameters to improve factual accuracy.
Why it's wrong here
Increasing model size can improve some capabilities but does not guarantee a reduction in hallucinations. Larger models may still generate plausible but incorrect information, especially if not grounded in verified sources. In a clinical context, relying solely on scale is insufficient and could even amplify confident errors, making it a poor mitigation strategy.
- ✗
Fine-tune the LLM on the company's historical clinical notes.
Why it's wrong here
Fine-tuning on historical notes may introduce biases and does not inherently reduce hallucinations. If the historical notes contain errors, the model could learn and perpetuate them. Moreover, fine-tuning alone does not provide real-time verification against authoritative sources, making it less effective for mitigating hallucinations in a high-stakes clinical environment.
About these practice questions
One of 367 original NCA-GENL practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official NVIDIA exam blueprint
This NCA-GENL practice question is part of Courseiva's free NVIDIA certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the NCA-GENL exam.