Understanding Azure OpenAI Content Filter Severity Thresholds
Refer to the exhibit.
{
"content_filters": [
{
"type": "hate",
"action": "block",
"severity": "high"
},
{
"type": "sexual",
"action": "block",
"severity": "medium"
},
{
"type": "self_harm",
"action": "block",
"severity": "low"
}
]
}You deploy an Azure OpenAI model with the above content filter configuration. A user submits a prompt that the system rates as "hate" at severity level "medium". What happens?
Quick Answer
Reading this correctly means treating each content filter category as its own independent threshold rather than a single blanket setting, since the configuration in the exhibit sets a different severity threshold for each category: hate is configured to block only at high, sexual content blocks at medium, and self-harm blocks at low. The prompt in question was rated hate at medium severity, and because the hate filter's threshold is specifically set to high, medium severity doesn't meet or exceed that bar — so the prompt is allowed through, even though the same medium rating would have been blocked if it had been classified under the sexual content category instead, which has a lower threshold. This is the detail these content-filter questions are testing: severity thresholds are per-category, not global, so the same detected severity level can produce a different outcome depending on which category triggered it. The action only fires when the detected severity meets or exceeds the threshold configured for that specific category, so correctly answering these questions means checking the detected category against its own row in the configuration, not assuming one threshold applies everywhere.
⚠ Common exam trap
A common mistake in Azure exams is assuming that any detection of a content type (e.g., hate) automatically triggers the configured action, ignoring that only severity levels meeting or exceeding the threshold are blocked.
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 prompt is allowed because the severity is below the threshold.
The content filter configuration blocks 'hate' content only at severity level 'high'. Since the user's prompt was rated as 'hate' at severity 'medium', it falls below the configured threshold and is allowed. Azure OpenAI content filters evaluate severity levels (low, medium, high) and apply the configured action only when the detected severity meets or exceeds the specified threshold.
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 prompt is allowed because the severity is below the threshold.
Why this is correct
The hate filter blocks only at high severity.
- ✗
The prompt is blocked because hate content is detected.
Why it's wrong here
Only high severity hate is blocked; medium is allowed.
- ✗
The prompt is blocked because the severity is medium.
Why it's wrong here
Block action is only for high severity.
- ✗
The prompt is allowed because the hate filter is not configured for medium.
Why it's wrong here
The filter is configured but with a higher threshold.
Go deeper
Related to this question
About these practice questions
This AI-102 question is part of Courseiva's 945-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 →
Same concept, more angles
1 more way this is tested on AI-102
These questions test the same concept from different angles. Work through them to make sure you can recognise it however the exam phrases it.
Variation 1. Which TWO statements about Azure OpenAI Service content filters are true?
easy- ✓ A.They can be configured with severity levels (low, medium, high)
- B.They only filter the output of the model
- C.They cannot be customized for specific use cases
- D.They are bypassed when using PTU deployments
- ✓ E.They include categories such as hate, sexual, violence, and self-harm
Why A: Azure OpenAI Service content filters can be configured with severity levels (low, medium, high) to control the strictness of filtering for each content category. This allows administrators to fine-tune the filter sensitivity based on their application's risk tolerance and compliance requirements.
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AI-102 practice question is part of Courseiva's free Microsoft 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 AI-102 exam.