Plan and manage an Azure AI solution →mediumMultiple ChoiceObjective-mapped

AI-102 Plan and manage an Azure AI solution Practice Question

Exhibit

{
  "features": {
    "search": {
      "type": "azure_ai_search",
      "index_name": "support-index",
      "semantic_configuration": "default"
    },
    "chat": {
      "type": "azure_openai",
      "deployment": "gpt-4o",
      "system_message": "You are a helpful assistant."
    },
    "content_safety": {
      "type": "azure_content_safety",
      "blocklist": "custom-blocklist",
      "threshold": "high"
    }
  }
}

Refer to the exhibit. You are configuring an Azure AI Foundry agent for customer support. The agent uses Azure AI Search for retrieval and Azure OpenAI for generation. Users report that the agent provides correct answers but sometimes includes inappropriate language. What is the most likely cause?

⚠ Common exam trap

Candidates may assume that content safety filtering is automatically applied to all outputs from Azure OpenAI, including those from AI Foundry agents. However, for agents, the blocklist must be explicitly assigned to the chat output in the agent configuration. Without this, the agent can generate inappropriate language even if the underlying model supports content filtering.

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

✓

The content safety blocklist is not applied to the chat output

Azure AI Foundry agents rely on Azure OpenAI content filtering to block harmful language in generated responses. If the content safety blocklist is not applied to the chat output, the agent can produce inappropriate language even when the retrieved information is accurate. The blocklist must be explicitly configured to filter the final output of the agent.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

✓
The content safety blocklist is not applied to the chat output
Why this is correct
Content safety configuration in the JSON is for input but may not be applied to the generated output; need to configure output filtering.
✗
The OpenAI deployment 'gpt-4o' does not support content filtering
Why it's wrong here
Azure OpenAI deployments support content filtering regardless of model.
✗
The content safety threshold should be set to 'low' to block more content
Why it's wrong here
'High' is the most restrictive; 'low' would allow more content.
✗
The semantic configuration is not optimized for safety
Why it's wrong here
Semantic configuration affects relevance, not content safety.

Go deeper

AI-102 Plan and manage an Azure AI solution Practice Question

Exhibit

Why each option matters

Option-by-option breakdown

Related to this question