VISIONAn Azure Content Moderator alternative, before March 2027.
Microsoft deprecated Azure Content Moderator in February 2024 and retires it on 15 March 2027. If your content is in Turkish or English, here is where each of its results goes in Detectivision, and what the code looks like before and after.
What Microsoft has announced
From Microsoft Learn, as read in October 2026. Detectivision is not affiliated with Microsoft.
Deprecated since February 2024
Azure Content Moderator has been deprecated since February 2024.
Retired on 15 March 2027
Microsoft will retire the service on 15 March 2027.
Microsoft’s replacement
Microsoft names Azure AI Content Safety as its replacement. Detectivision is another way out, for content in Turkish and English.
Where each Content Moderator result goes
Content Moderator returns scores and matched terms; Detectivision raises flags, each with its confidence.
| Content Moderator | Detectivision | Note |
|---|---|---|
| Text Category1 | sexual | Language that might be sexually explicit or adult. |
| Text Category2 | sexual | Sexually suggestive or mature language: the same flag as Category1. |
| Text Category3 | insult, hate, racism, sexism | Offensive language, one flag per kind. |
| ReviewRecommended | flagged | True when any flag you asked for was raised. |
| Terms (profanity) | none | Not among the flags every account has. Swearing aimed at someone is an insult. |
| PII | none | Not among the flags every account has. |
| Image adult and racy | sexual | One image flag for explicit and suggestive images. |
| OCR and faces | none | No flag of their own. |
| Custom term and image lists | none | Keep your own lists in your code. |
Some flags have no Content Moderator result: self_harm and bad_habits (drug use, drinking, smoking or gambling talked up) in text, and violence, hate and scam in images.
A text check, both ways
curl -X POST \
"$AZURE_ENDPOINT/contentmoderator/moderate/v1.0/ProcessText/Screen?classify=True" \
-H "Ocp-Apim-Subscription-Key: $AZURE_KEY" \
-H "Content-Type: text/plain" \
--data "You are an idiot"curl https://api.detectivision.ai/api/v1/moderation \
-H "Api-Key: $DETECTIVISION_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"type": "text_moderation",
"input": "You are an idiot",
"flags": ["insult", "hate", "racism", "sexism", "sexual"]
}'Content Moderator classifies English only. Detectivision checks Turkish and English, from the text in input, with no language to set.
An image check, both ways
curl -X POST \
"$AZURE_ENDPOINT/contentmoderator/moderate/v1.0/ProcessImage/Evaluate" \
-H "Ocp-Apim-Subscription-Key: $AZURE_KEY" \
-H "Content-Type: application/json" \
-d '{ "DataRepresentation": "URL", "Value": "https://example.com/photo.jpg" }'curl https://api.detectivision.ai/api/v1/moderation \
-H "Api-Key: $DETECTIVISION_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"type": "image_moderation",
"input": "https://example.com/photo.jpg",
"mode": "detect",
"flags": ["sexual"]
}'isImageAdultClassified and isImageRacyClassified become one flag, sexual, with its confidence. detect mode answers with the verdict only. The image is not stored.
What is different
Languages
Detectivision checks Turkish and English. Content Moderator classifies English only, and matches profane terms in many languages.
Price
Paid in credits: an image is 1 credit in detect mode and a text 1 credit per started 2,500 characters, by webhook; twice that at once.
More flags
Violence, hate and scams in images; self-harm and bad habits in text.
Actions
Moderators act on each request from the dashboard’s Logs, and your system gets every action by webhook.
Async
Content that can wait is sent with "async": true, at half the price; a burst waits its turn instead of being refused.
Turkish
Insults, hate, racism, sexism and sexual language in Turkish text.
Frequently asked
When is Azure Content Moderator retired?
Microsoft says it is deprecated as of February 2024 and will be retired on 15 March 2027.
What does Microsoft recommend instead?
Azure AI Content Safety. Detectivision is another option for content in Turkish and English.
Is there a ReviewRecommended in Detectivision?
flagged plays that part: it is true when any of your flags was raised. Each flag has its own confidence, so you can draw your own line.
Does Detectivision find personal data or match term lists?
Not with the flags every account has. GET /api/v1/flags lists the flags of your account.
Which languages does Detectivision support?
Turkish and English. Text in other languages is checked, but the results are not reliable.
Can I try it on my own content first?
Yes. Start the free trial, then send your texts and images through the dashboard’s playground or the API.