Content filter

At a glance

The community member is experiencing a content-management error with the AzureChatOpenAI API, receiving an error message indicating that the prompt has triggered Azure OpenAI's content management policy. The community members suggest that the word "bullet" may have triggered the content filter, and that the issue is specific to the Azure API. They also mention a related GitHub issue and suggest trying to upgrade the OpenAI library, but the community member has the latest version. There is no explicitly marked answer in the comments.

Useful resources

ssnapster

I'm hitting content-management error with AzureChatOpenAI

Plain Text

APIError: Invalid response object from API: '{"error":{"message":"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","type":null,"param":"prompt","code":"content_filter","status":400}}' (HTTP response code was 400)

Need some help

Attachment

8 comments

LLogan M

I'm not totally sure here, I'm not an expert with azure.

Just as a test, does it work with use_async=False?

LLogan M

Ohhhh it's probably because you used the word bullet actually ... seems like that might trigger the content filter 😅 azure must be very strict

ssnapster

@Logan M i got this even with query ""give me top local news around US.""

ssnapster

can it be somehow related to this https://github.com/openai/openai-python/issues/306 ?

LLogan M

hmmm, I'm not sure then. I haven't seen this before actually, at least on discord.

Maybe try pip install --upgrade openai, judging by that github issue?

ssnapster

have the latest openai

ssnapster

this is strictly related only to azure API right? probably why you guys haven't seen this before

LLogan M

Yea looks like it should only happen for azure 🤔😕

Add a reply

Find answers from the community

Content filter