A team has deployed an LLM-based content moderation tool with an output guardrail that flags toxic language. After a month, they notice that the guardrail is missing new slang terms that are being used to bypass it. They want to continuously improve the guardrail. Which approach is most effective?

Remove the guardrail and rely on human moderation.

Increase the sensitivity of the guardrail to flag more content.

Regularly update the guardrail's detection rules based on new examples of bypass attempts.

Replace the guardrail with a more advanced model that understands slang.

A company is building a chatbot that must comply with data privacy regulations. They want to ensure that the chatbot does not reveal personal information in its responses. Which guardrail approach is most appropriate?

An input guardrail that blocks any prompt containing personal information.

Disable the chatbot's ability to access external data.

A system prompt that instructs the model not to reveal personal information.

An output guardrail that detects and redacts personal information from the model's responses.

Which practice best supports continuous improvement of guardrails?

Only using guardrails that are built into the model

Setting guardrails once and never changing them

Regularly reviewing guardrail performance on new test cases

Evaluating guardrails only after a major incident

DatabricksCertified Generative AI Engineer Associate

Domain 3Objective 4

Safety and Guardrails GENERATIVE-AI-ENGINEER-ASSOCIATE Practice Questions (Page 1)

Part of the Section 3: Application Development domain, which makes up ~18% of our current practice bank. Databricks does not publish an official question count, but from its 90-minute exam (~35–60 total, ~6–11 in this domain), expect 1–2 from this objective — we provide 31 practice questions to prepare you well beyond it. (estimate)

31questions here

7free pages

7concepts

Questions 1–5

1expert · hard
A team has deployed an LLM-based content moderation tool with an output guardrail that flags toxic language. After a month, they notice that the guardrail is missing new slang terms that are being used to bypass it. They want to continuously improve the guardrail. Which approach is most effective?
Select an answer first
2foundation · easy
After an LLM generates a response, a guardrail checks whether the response contains any toxic language and, if so, replaces it with a safe message. This is an example of which type of guardrail?
Select an answer first
3application · medium
A company is building a chatbot that must comply with data privacy regulations. They want to ensure that the chatbot does not reveal personal information in its responses. Which guardrail approach is most appropriate?
Select an answer first
4foundation · easy
Which practice best supports continuous improvement of guardrails?
Select an answer first
5application · medium
During a qualitative review of an LLM's responses, a tester notes that the model sometimes produces responses that are grammatically correct but contain contradictions within the same paragraph. Which type of issue is this?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to GENERATIVE-AI-ENGINEER-ASSOCIATE

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by Databricks. “GENERATIVE-AI-ENGINEER-ASSOCIATE” is a trademark of its owner, used for identification only.

Safety and Guardrails GENERATIVE-AI-ENGINEER-ASSOCIATE Practice Questions (Page 1)

Questions 1–5

A team has deployed an LLM-based content moderation tool with an output guardrail that flags toxic language. After a month, they notice that the guardrail is missing new slang terms that are being used to bypass it. They want to continuously improve the guardrail. Which approach is most effective?

After an LLM generates a response, a guardrail checks whether the response contains any toxic language and, if so, replaces it with a safe message. This is an example of which type of guardrail?

A company is building a chatbot that must comply with data privacy regulations. They want to ensure that the chatbot does not reveal personal information in its responses. Which guardrail approach is most appropriate?

Which practice best supports continuous improvement of guardrails?

During a qualitative review of an LLM's responses, a tester notes that the model sometimes produces responses that are grammatically correct but contain contradictions within the same paragraph. Which type of issue is this?

Keep studying — Section 3: Application Development