Before launching a public-facing Foundry-hosted assistant on its website, a retailer's security team wants to reduce the chance that the deployment is exploited for abusive purposes and wants ongoing visibility if abuse happens after launch. Which two actions should they take? (Select TWO.)
Select all correct answers, then click Submit.
Short Explanation
There are really two separate jobs here: stopping bad stuff before it happens, and knowing about it if it happens anyway. Blocking harmful content on both what goes in and what comes out handles the first job, catching problems at the door. Turning on monitoring with alerts handles the second job, giving the team a way to notice a problem after launch instead of finding out from a news story. Making every response identical just removes variety, it does not screen for anything harmful. And turning off logging is actually the opposite of what you want here, since without logs there is nothing to look back at when something does get flagged. The two moves that actually address prevention and after-the-fact visibility are filtering the content and watching for abuse patterns.
Full Explanation
The correct answers are A and C. Configuring content filters on both prompts and completions blocks harmful categories before they reach the model or before a harmful response reaches the user, directly reducing the chance of exploitation. Enabling abuse monitoring with alerts gives the team ongoing visibility after launch, notifying them when usage patterns look abusive so they can investigate and respond quickly, which is exactly the second requirement stated in the scenario. Option B is incorrect because setting temperature to zero only makes responses more deterministic and repeatable; it has no effect on detecting or preventing abusive prompts or outputs. Option D is incorrect because disabling logging removes the very data needed to investigate flagged usage and understand what happened during a suspected abuse event, which works directly against the team's goal of ongoing visibility rather than supporting it. Together, content filtering and abuse monitoring address both halves of the requirement: prevention at the point of interaction, and detection and follow-up over time, which is why these two, and not the other two options, form the correct combination for this pre-launch checklist.