[BUG] Azure OpenAI - Filtering bad request

[ZaneBartlett1]

Library name and version

Azure.AI.OpenAI Version=2.1.0-beta.1

Describe the bug

I do not have an easy way to see what the content filtering type was that filtered my message from OpenAI. At this link https://learn.microsoft.com/en-us/azure/ai-services/openai/concepts/content-filter?tabs=warning%2Cuser-prompt%2Cpython-new they show

{
    "error": {
        "message": "The response was filtered due to the prompt triggering Azure Content management policy. 
                   Please modify your prompt and retry. To learn more about our content filtering policies
                   please read our documentation: https://go.microsoft.com/fwlink/?linkid=21298766",
        "type": null,
        "param": "prompt",
        "code": "content_filter",
        "status": 400,
        "innererror": {
            "code": "ResponsibleAIPolicyViolation",
            "content_filter_result": {
                "hate": {
                    "filtered": true,
                    "severity": "high"
                },
                "self-harm": {
                    "filtered": true,
                    "severity": "high"
                },
                "sexual": {
                    "filtered": false,
                    "severity": "safe"
                },
                "violence": {
                    "filtered":true,
                    "severity": "medium"
                }
            }
        }
    }
}

When I hit a content filtering response I get this as the message

HTTP 400 (content_filter)
Parameter: prompt

The response was filtered due to the prompt triggering Azure OpenAI's content management policy. Please modify your prompt and retry. To learn more about our content filtering policies please read our documentation: https://go.microsoft.com/fwlink/?linkid=2198766"

but all other properties are not exposed. I see that in the response I get back there is InnerExpection but it is null everytime, and I believe this is a bug. When I dig deeper in the debug console into non-public members, I see that innerError exists, but because it's non-public I have to use reflect to access it to log out the content filter.

Expected behavior

That more properties from the actual error message would make it into exposed properties, particularly, that innererror would be exposed as innerExpection

{
    "error": {
        "message": "The response was filtered due to the prompt triggering Azure Content management policy. 
                   Please modify your prompt and retry. To learn more about our content filtering policies
                   please read our documentation: https://go.microsoft.com/fwlink/?linkid=21298766",
        "type": null,
        "param": "prompt",
        "code": "content_filter",
        "status": 400,
        "innererror": {
            "code": "ResponsibleAIPolicyViolation",
            "content_filter_result": {
                "hate": {
                    "filtered": true,
                    "severity": "high"
                },
                "self-harm": {
                    "filtered": true,
                    "severity": "high"
                },
                "sexual": {
                    "filtered": false,
                    "severity": "safe"
                },
                "violence": {
                    "filtered":true,
                    "severity": "medium"
                }
            }
        }
    }
}

Actual behavior

innerExpection is null everytime and the innererror is left out as a non-public member. This is basically all the useful information thta I'm getting out from the exception

{System.ClientModel.ClientResultException: HTTP 400 (content_filter)
Parameter: prompt

The response was filtered due to the prompt triggering Azure OpenAI's content management policy. Please modify your prompt and retry. To learn more about our content filtering policies please read our documentation: https://go.microsoft.com/fwlink/?linkid=2198766

Reproduction Steps

Create a generic try catch exception where the try makes a call using the CompleteChatAsync method and an input that would get flagged from the content filters. Try to log out innererror in the catch section somehow not using reflection and realize you can not.

Environment

No response

[github-actions]

Thanks for the feedback! We are routing this to the appropriate team for follow-up. cc @jpalvarezl @ralph-msft @trrwilson.