Databricks-DA-Assoc Developing AI/BI Genie Spaces Practice Question
When designing an AI/BI Genie Space, how can an analyst prevent the model from accessing PII (Personally Identifiable Information) columns in a sensitive table?
⚠ Common exam trap
Candidates mistakenly believe they can hide PII columns using Genie system instructions, ignoring the fact that prompt instructions can be bypassed or fail at the data layer.
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
✓
Create a filtered view in Unity Catalog that excludes sensitive columns.
The best approach to prevent PII exposure is to create a secure view in Unity Catalog that excludes the PII columns and point the Genie Space to that view instead of the raw table. This effectively enforces data masking and access control at the data layer. By only exposing the non-sensitive data required for analysis, you maintain privacy compliance while still allowing the AI to generate useful insights for the business.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Add a disclaimer in the system instructions to 'not use PII'.
Why it's wrong here
Relying on an instruction to 'not use PII' is insufficient from a security perspective. It is a 'soft' control that the model might ignore. For true data protection, you must implement 'hard' controls at the data layer, such as row-level security or column-level redaction in the view definition.
- ✗
Use a specialized, private LLM that has no knowledge of PII.
Why it's wrong here
Even if the model has no knowledge of PII, it can still surface sensitive data contained in the database if it has access to the tables. Security must be managed at the data level, not by hoping the model is 'unaware' of the data it is accessing and processing.
- ✓
Create a filtered view in Unity Catalog that excludes sensitive columns.
Why this is correct
This is the most secure approach. By creating a view that explicitly selects only the non-sensitive columns, you guarantee that the LLM cannot 'see' or include PII in its SQL generation. This fulfills data governance requirements and protects sensitive information regardless of what the user asks the model.
- ✗
Set the temperature to 0 to minimize hallucinations involving PII.
Why it's wrong here
Lowering the temperature does not prevent the model from accessing existing, authorized data columns. If the model has access to the table, it can still query PII columns. You must remove access at the database level to ensure that sensitive information is physically excluded from the model's available dataset.
About these practice questions
This Databricks-DA-Assoc question is part of Courseiva's 291-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam 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.