Skip to content

Commit eafa040

Browse files
authored
Enhance ZDR Overview with data verification steps
Added details on verifying data storage for abuse monitoring in Azure.
1 parent 53b10ba commit eafa040

1 file changed

Lines changed: 32 additions & 3 deletions

File tree

0_Azure/3_AzureAI/9_AzureOpenAI/demos/12_ZDR_Overview.md

Lines changed: 32 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -14,15 +14,44 @@ Last updated: 2025-12-03
1414

1515
</details>
1616

17-
1817
> ZDR - Microsoft is committed to not retaining any customer requests or response data beyond what is strictly necessary for real-time processing. Once the data has fulfilled its intended purpose, it is promptly deleted, ensuring that prompts and completions are not stored for training, debugging, or analytics purposes.
1918
20-
2119
<img width="943" height="1052" alt="flow-3" src="https://github.com/user-attachments/assets/efe3ce83-75f5-4567-8ab3-384bb61fff36" />
2220

23-
2421
From [How does Foundry process data to provide Azure Direct Models?](https://learn.microsoft.com/en-us/azure/ai-foundry/responsible-ai/openai/data-privacy?view=foundry-classic&tabs=azure-portal#how-does-foundry-process-data-to-provide-azure-direct-models)
2522

23+
## How can a customer verify if data storage for abuse monitoring is off?
24+
25+
> Once approval is granted, data storage for abuse monitoring will be disabled. You can confirm this status using the Azure portal or Azure CLI. For further details, please consult the documentation here: [How can a customer verify if data storage for abuse monitoring is off?](https://learn.microsoft.com/en-us/azure/ai-foundry/responsible-ai/openai/data-privacy?view=foundry-classic&tabs=azure-portal#how-can-a-customer-verify-if-data-storage-for-abuse-monitoring-is-off)
26+
27+
> Using [Azure CLI](https://learn.microsoft.com/en-us/azure/ai-foundry/responsible-ai/openai/data-privacy?view=foundry-classic&tabs=azure-cli#prerequisites):
28+
29+
```azurecli
30+
az cognitiveservices account show -n resource\_name -g resource \_group
31+
```
32+
33+
> [!NOTE]
34+
> The system will return a JSON response. If abuse monitoring logging is disabled, you will find a value labeled `ContentLogging`, which will be set to `FALSE`.
35+
36+
37+
> Using [Azure Portal](https://learn.microsoft.com/en-us/azure/ai-foundry/responsible-ai/openai/data-privacy?view=foundry-classic&tabs=azure-portal#prerequisites):
38+
39+
For example: In my Azure OpenAI instance, I'm not able to find the `ContentLogging` attribute. `This attribute is only visible when set to false, indicating that abuse monitoring has been disabled.` In this case, for me is on.
40+
41+
<img width="1911" height="830" alt="image" src="https://github.com/user-attachments/assets/35f43b1b-35cb-49f0-bf57-048575cbd6d5" />
42+
43+
<img width="1883" height="827" alt="image" src="https://github.com/user-attachments/assets/4814d8e9-84fd-485a-8f8f-03126d09bf2a" />
44+
45+
> When it is set to `off`, you should be able to view the following:
46+
47+
```
48+
{
49+
"name":"ContentLogging",
50+
"value":"false"
51+
}
52+
```
53+
54+
2655

2756
<!-- START BADGE -->
2857
<div align="center">

0 commit comments

Comments
 (0)