Forum Widgets
Latest Discussions
Azure OpenAI Content Filter Result is always content_filter_error
I'm exploring blocklists as a solution for OpenAI not detecting sensitive words (specifically "wrist-cutting" in my local language (Cantonese) (to be fair not even Chinese AIs know the word) I have created a Blocklist with 1 entry: Term: [鎅𰾛𠝹]手 Type: Regex It can block inputs with ease: { "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, "innererror": { "code": "ResponsibleAIPolicyViolation", "content_filter_result": { "custom_blocklists": { "details": [ { "filtered": true, "id": "ChineseBlockList" } ], "filtered": true }, "hate": { "filtered": false, "severity": "safe" }, "profanity": { "filtered": false, "detected": false }, "self_harm": { "filtered": false, "severity": "safe" }, "sexual": { "filtered": false, "severity": "safe" }, "violence": { "filtered": false, "severity": "safe" } } } } } However, it cannot block outputs. { "choices": [ { "content_filter_result": { "error": { "code": "content_filter_error", "message": "The contents are not filtered" } }, "content_filter_results": {}, "finish_reason": "stop", "index": 0, "logprobs": null, "message": { "content": "𠝹手（也寫作“拍手”）是一種手部動作，通常是將雙手合攏並用力拍打在一起，發出聲音。這個動作常用於表達讚賞、鼓勵或慶祝，像是在演出結束後觀眾的掌聲，或是在某些活動中用來引起注意。𠝹手也可以用於節奏感的表達，像是在音樂中隨著節拍拍手。這個動作在許多文化中都有其獨特的意義和用途。", "refusal": null, "role": "assistant" } } ], "created": 1737702254, "id": "chatcmpl-At81eUTIzDkZPCKznSKr19YMJU1ud", "model": "gpt-4o-mini-2024-07-18", "object": "chat.completion", "prompt_filter_results": [ { "prompt_index": 0, "content_filter_results": { "custom_blocklists": { "filtered": false }, "hate": { "filtered": false, "severity": "safe" }, "profanity": { "filtered": false, "detected": false }, "self_harm": { "filtered": false, "severity": "safe" }, "sexual": { "filtered": false, "severity": "safe" }, "violence": { "filtered": false, "severity": "safe" } } } ], "system_fingerprint": "fp_5154047bf2", "usage": { "completion_tokens": 138, "completion_tokens_details": { "accepted_prediction_tokens": 0, "audio_tokens": 0, "reasoning_tokens": 0, "rejected_prediction_tokens": 0 }, "prompt_tokens": 34, "prompt_tokens_details": { "audio_tokens": 0, "cached_tokens": 0 }, "total_tokens": 172 } }
angular-b2c-sample-app and iframes
I have a project that I based on this sample project: https://github.com/AzureAD/microsoft-authentication-library-for-js/tree/dev/samples/msal-angular-v3-samples/angular-b2c-sample-app Currently I have the custom policies as intended. Button, redirects to a azure page, finishes w/e the custom policy is, redirects back to the web app. But now I need to using said custom policy as an iframe, basically the client requests that there is less one layer of buttons to pressed. Therefore I want the custom policy to be displayed as an iframe, making the web app feel more single page. I have looked around, but can't really find anything detailed enough. Any help is welcomed :)
Entra Connect - Access to onsite Server Shares
Will setting up Entra Connect and syncing AD IDs with Entra IDs provide access to on Prem Windows server shares? We are planning for our migration to the cloud and have a bunch of Windows server shares that users currently access with their Active Directory accounts. Setting up Entra Connect and syncing the AD accounts to Entra ID Accounts provide the same access to these shares? Migrating the file shares to Entra ID is a bigger migration project probably further down the road. I found this article but it's the other way around, Hybrid identities accessing Azure shares: https://learn.microsoft.com/en-us/azure/storage/files/storage-files-identity-auth-hybrid-identities-enable?tabs=azure-portal%2CintuneNew Azure GxP guidelines help pharmaceutical and biotech customers build GxP solutions
We recently released a detailed set of GxP qualification guidelines for our Azure customers. These guidelines give life sciences organizations, such as pharmaceutical and biotechnology companies, a comprehensive toolset for building solutions that meet GxP compliance regulations. GxP is a general abbreviation for "good practice" quality guidelines and regulations. Technology systems that use GxP processes such as Good Laboratory Practices (GLP), Good Clinical Practices (GCP), and Good Manufacturing Practices (GMP) require validation of adherence to GxP. Solutions are considered qualified when they can demonstrate the ability to fulfill GxP requirements. GxP regulations include pharmaceutical requirements, such as those outlined in the U.S. Food and Drug Administration CFR Title 21 Part 11, and EU GMP Annex 11. Read about it in the Azure blog.Running self hosted agent in interactive mode on on-prem server
I have a scenario, Where I have some UFT Developer Automation Tests cases that carry UI automation of Facets Application. I have them running on TFS release pipelines but the test cases are failing because the agent is not able to find the Facets window as it is running as a service and we need to configure the ADO agent to run in interactive mode for UI test. Instead of manually logging in to the server and trigger the agent to run in interactive how can we automate that because i need 50 agents to run in the interactive mode
Error in Manager data synchronization via SCIM
Hello! My name is Alex Moiseev, and I am experiencing an issue with user synchronization via SCIM with Azure Active Directory. Well, actually with manager synchronization. As far as I understand, the approach of provisioning of the user is the following. Step 1. Provisioner tries to get info about the user by id with GET User resource request. Step 2. Based on information received, Provisioner decides should the whole information about the user be sent or there is a need only to update certain fields. Step 3. Provisioner send POST or PUT/PATCH request with user details to create/update user on the receiver side. And everything works more or less ok, but the managers. In user data to provide there is one field, which is used for manager info: urn:ietf:params:scim:schemas:extension:enterprise:2.0:User:manager When there is no manager on consumer side assigned to the user, everything works perfectly. The provisioner sends manager info in both cases - with existing and with non-existing user. If the manager is changed on Azure AD side, we still receive manager data in the field mentioned in PATCH request in order to overwrite stored manager. But when the manager is removed on Azure AD side, we didn't receive any information about it - urn:ietf:params:scim:schemas:extension:enterprise:2.0:User:manager field doesn't exist in PATCH request. We thought, that may be because we didn't add manager information in GET User resource response. We tried to add manager information there according to the documentation (https://docs.microsoft.com/en-us/azure/active-directory/app-provisioning/use-scim-to-provision-users-and-groups) by adding "urn:ietf:params:scim:schemas:extension:enterprise:2.0:User": { "manager": "0" }, We use "0" in order to make the provisioner to send us information about the manager in each request. But we've got an error there: Error message We are not able to deserialize the resource received from your SCIM endpoint because your SCIM endpoint is not fully compatible with the Azure Active Directory SCIM client. Here is the resource we received from your SCIM endpoint: { "schemas”: [ "urn:ietf:params:scim:schemas:core:2.0:User", "urn:ietf:params:scim:schemas:extension:enterprise:2.0:User" ], "externalId": "Worker", "id": "1548197", "userName": "email address removed for privacy reasons", "name": { "familyName": "Ker", "givenName": "Wor" }, "emails": [ { "value": "email address removed for privacy reasons", "type": "work", "primary": true } ], "title": "Developer", "locale": "nl", "timezone": "CEST", "urn:ietf:params:scim:schemas:extension:enterprise:2.0:User": { "manager": "0" }, "active": true, "displayName": "Wor Ker" } Please refer to the Azure Active Directory SCIM provisioning documentation (https://docs.microsoft.com/en-us/azure/active-directory/app-provisioning/use-scim-to-provision-users-and-groups) and adapt the SCIM endpoint to be able to process provisioning requests from Azure Active Directory. We have reviewed the documentation referenced, and the format for specifying the manager is exactly as indicated in the documentation. If the "manager" field is removed from the data, the error does not occur. So, the questions are: 1. How should we provide manager information in the response of GET User resource call? 2. How should we catch the removing of the manager in Azure in a proper way? I would appreciate your assistance in resolving this issue. I'd like to know what is causing the error and how to correctly transmit manager information via SCIM. Thank you in advance for your attention and help! Yours sincerely, Alex Moiseev
