Databricks Certified Generative AI Engineer Associate — Question 19

A Generative AI Engineer is developing an LLM application that users can use to generate personalized birthday poems based on their names.
Which technique would be most effective in safeguarding the application, given the potential for malicious user inputs?

Answer options

• A. Implement a safety filter that detects any harmful inputs and ask the LLM to respond that it is unable to assist
• B. Reduce the time that the users can interact with the LLM
• C. Ask the LLM to remind the user that the input is malicious but continue the conversation with the user
• D. Increase the amount of compute that powers the LLM to process input faster

Correct answer: A

Explanation

The correct answer is A because implementing a safety filter is a proactive measure to prevent harmful inputs from affecting the application. Option B does not address the issue of input safety, while option C allows harmful inputs to be acknowledged but doesn't prevent their impact. Option D focuses on performance rather than input security.