Configuring Azure OpenAI

Azure OpenAI can be integrated with AppScan via API endpoints to enhance the true positive rate of scan results by adding additional filters.

Follow these steps to configure Azure OpenAI and improve your scan results.

  1. Navigate to Configuration > Test options > Azure OpenAI configuration.
  2. Enter the Endpoint in the following format: {azure_openai_endpoint}/openai/deployments/{deployment_name}/completions
    Table 1. Endpoint elements
    Element Description Example
    {azure_openai_endpoint}

    Substitute with the value from the Endpoint field under the Keys & Endpoint section from the Azure portal.

    https://aoairesource.openai.azure.com
    {deployment_name} Corresponds to the custom name you chose for your deployment when you deployed a model. You can get this value from Azure OpenAI Studio of your Azure portal.
    Note: Azure OpenAI testing was conducted using GPT-4o, version 2024-08-06. We recommend using this specific model and version to avoid any failures.
    GPT-4o
    Example of endpoint: https://aoairesource.openai.azure.com/openai/deployments/GPT-4o/completions
  3. Enter the API Key. This value can be found in the Keys & Endpoint section when examining your resource from the Azure portal. You can use either KEY1 or KEY2.
    CAUTION: Ensure both the Endpoint and API Key values are correct, as there is no validation to catch typos or errors.
  4. Click Apply to save the configuration.

    Result: The Azure OpenAI endpoint and API key have been configured.

What to do next: Start your scan in AppScan as usual.