How to remove content filtering from Azure Open AI Deployment

Question

How to remove content filtering from Azure Open AI Deployment

Marlon Sanguino 0

Hello,

I would like to ask for clarification regarding the removal of content filtering in an OpenAI deployment.

I currently have two default profiles and have created one with all settings set to "low." However, I would like to completely remove the content filtering due to the nature of the application I'm using. I have already submitted a request for this and was informed by the Microsoft team that it was approved, but I am unsure how to confirm whether the content filter has actually been removed. Additionally, I would appreciate guidance on how to configure the deployment without any filtering.

Could you please provide instructions or confirm the current status of the content filter in my deployment?

Thanks,

Sina Salam 22,031 Reputation points Volunteer Moderator

2024-10-15T16:58:30.1833333+00:00
Hello Marlon Sanguino,

Welcome to the Microsoft Q&A and thank you for posting your questions here.

I understand that you would like to affirm on how you can remove content filtering from Azure Open AI Deployment.

If you need to configure the deployment without any filtering, do the following steps:

Go to your Azure OpenAI Service deployment settings.

Navigate to the content filtering section.

Set all content filtering options to "low" or disable them entirely.

Read more in this link: https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/content-filters

I hope this is helpful! Do not hesitate to let me know if you have any other questions.

Please don't forget to close up the thread here by upvoting and accept it as an answer if it is helpful.
Marlon Sanguino 0 Reputation points

2024-10-15T20:06:38.04+00:00

Hello Sina,

Thank you for your prompt response.

However, in my filter settings, I do not have the option to disable the filter entirely. Please see the attached image for reference.

I have also submitted a request to Cognitive Services Gating Support, and they informed me that it was approved. However, I'm unsure if that is accurate or how I can fully disable the content filtering from my model deployment.

Thank you for your assistance.
Sina Salam 22,031 Reputation points Volunteer Moderator

2024-10-15T20:35:18.1966667+00:00
Hi Marlon Sanguino

Thank you for your feedback.

Set all content filtering options to "low" or disable them entirely.

if you don't see disable use low to all content filtering.

cheers
Marlon Sanguino 0 Reputation points

2024-10-15T21:49:28.4466667+00:00

Hi Sina,

The issue is that even with the filter set to "Low," it is still filtering content that we do not want filtered. We would like to completely disable the content filtering altogether.

Thank you for your help.
giangpham 0 Reputation points

2024-12-06T13:05:11.3966667+00:00
This does not help Marlon!

No disable available

Set to low mean it block completely

Set to High it show it wil block Low and Medium

But all are very bad with very common question.

3 answers

Your answer

Sina Salam 22,031 Reputation points Volunteer Moderator

2024-10-15T16:58:30.1833333+00:00

Hello Marlon Sanguino,

Welcome to the Microsoft Q&A and thank you for posting your questions here.

I understand that you would like to affirm on how you can remove content filtering from Azure Open AI Deployment.

If you need to configure the deployment without any filtering, do the following steps:

Go to your Azure OpenAI Service deployment settings.

Navigate to the content filtering section.

Set all content filtering options to "low" or disable them entirely.

Read more in this link: https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/content-filters

I hope this is helpful! Do not hesitate to let me know if you have any other questions.

Please don't forget to close up the thread here by upvoting and accept it as an answer if it is helpful.
Marlon Sanguino 0 Reputation points

2024-10-15T20:06:38.04+00:00

Hello Sina,

Thank you for your prompt response.

However, in my filter settings, I do not have the option to disable the filter entirely. Please see the attached image for reference.

I have also submitted a request to Cognitive Services Gating Support, and they informed me that it was approved. However, I'm unsure if that is accurate or how I can fully disable the content filtering from my model deployment.

Thank you for your assistance.
Sina Salam 22,031 Reputation points Volunteer Moderator

2024-10-15T20:35:18.1966667+00:00

Hi Marlon Sanguino

Thank you for your feedback.

Set all content filtering options to "low" or disable them entirely.

if you don't see disable use low to all content filtering.

cheers
Marlon Sanguino 0 Reputation points

2024-10-15T21:49:28.4466667+00:00

Hi Sina,

The issue is that even with the filter set to "Low," it is still filtering content that we do not want filtered. We would like to completely disable the content filtering altogether.

Thank you for your help.
giangpham 0 Reputation points

2024-12-06T13:05:11.3966667+00:00

This does not help Marlon!

No disable available

Set to low mean it block completely

Set to High it show it wil block Low and Medium

But all are very bad with very common question.

Answer 1

Deleted

This answer has been deleted due to a violation of our Code of Conduct. The answer was manually reported or identified through automated detection before action was taken. Please refer to our Code of Conduct for more information.

Comments have been turned off. Learn more

Answer 2

Omar 0

I don't see a way to disable it either.

I am re-deploying with different content safety setting (to none/off).

Would be good to be able to toggle this more easily.

Answer 3

Wayne Bae 0 Microsoft Employee

I think the only option that you can choose to modify the content filtering is by editing the deployment properties in your Azure AI Foundry -> Deployments -> Your model -> Edit -> Content Filter.

You can choose among a few options there. In my case, I switched to Default instead of DefaultV2 (default option) since the first one doesn't have jailbreak filter, which I always got during my test even without any attempt to jailbreak the model.

Share via

How to remove content filtering from Azure Open AI Deployment

3 answers

Your answer