Databricks-DA-Assoc Developing AI/BI Genie Spaces Practice Question
Which TWO actions should an analyst take to improve the accuracy of an AI/BI Genie Space when users report that it frequently hallucinates metric definitions?
⚠ Common exam trap
Test-takers often select only one corrective action, forgetting that stopping hallucinations requires both clear natural language instructions and curated Unity Catalog views.
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
✓
Provide clear, concise system instructions that define key metrics.
Improving AI/BI Genie accuracy involves a dual approach: tightening the instruction set and refining the data layer. By providing specific, clear instructions, the analyst reduces ambiguity. Simultaneously, creating curated views in Unity Catalog enforces a 'source of truth,' limiting the model's ability to interpret ambiguous raw columns. These steps collectively ground the LLM, ensuring that it generates reliable SQL based on verified organizational definitions rather than probabilistic guesses.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Increase the number of tables available to the Genie Space.
Why it's wrong here
Adding more tables increases the search space for the LLM, often leading to more confusion and higher hallucination rates. Instead of breadth, the goal should be depth and precision. A focused, curated set of tables is significantly better for accuracy than a wide, unmanaged collection of raw data sources.
- ✓
Provide clear, concise system instructions that define key metrics.
Why this is correct
Well-defined system instructions provide the necessary 'guardrails' for the LLM. By explicitly stating how metrics should be calculated, the developer forces the model to follow specific business logic. This drastically reduces the likelihood of the model creating its own interpretations or using incorrect formulas for standard business KPIs.
- ✓
Create standardized views in Unity Catalog and point Genie to those.
Why this is correct
Standardized views act as a semantic layer that abstracts complexity. By exposing only the necessary columns and calculated fields, you prevent the LLM from accessing raw, potentially misleading data. This forces the model to use the pre-calculated, verified logic defined in the SQL view, leading to more consistent results.
- ✗
Enable 'auto-discovery' of all columns in the Unity Catalog tables.
Why it's wrong here
Auto-discovery can expose sensitive or irrelevant columns that confuse the model. It often leads to the inclusion of internal identifiers or metadata fields in analytical queries. Controlled, intentional exposure of data is a best practice for maintaining accuracy and security in any AI-driven analytical application or environment.
- ✗
Use a higher temperature setting to allow for more flexible interpretations.
Why it's wrong here
Higher temperatures make the model less predictable and more prone to generating varied, potentially incorrect SQL queries. For business intelligence applications, where correctness is paramount, low temperature settings are essential. You want the model to be as deterministic as possible when it comes to standard business metric calculations.
About these practice questions
Courseiva writes every Databricks-DA-Assoc question from scratch — 291 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. 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 Databricks exam blueprint
This Databricks-DA-Assoc practice question is part of Courseiva's free Databricks 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 Databricks-DA-Assoc exam.