@@ -3,17 +3,16 @@ author: laujan
manager: nitinme
ms.service: azure-ai-language
ms.topic: include
-ms.date: 02/06/2026
+ms.date: 04/27/2026
ms.author: lajanuar
-ms.custom: language-service-pii
-ai-usage: ai-assisted
+ms.custom: doc-kit-assisted
---
## Prerequisites
> [!TIP]
>
-> * If you already have an Azure Language in Foundry Tools or multi-service resource—whether used on its own or through Language Studio—you can continue to use those existing Language resources within the Microsoft Foundry portal.
+> * If you already have an Azure Language in Foundry Tools or multi-service resource—whether used on its own or through Language Studio—you can continue to use those existing Language resources within the Microsoft Foundry portal.
> * For more information, see [Connect services in the Microsoft Foundry portal](../../../../connect-services-foundry-portal.md).
> * Consider using a Foundry resource for the best experience. You can also follow these instructions with a Language resource.
@@ -22,171 +21,210 @@ ai-usage: ai-assisted
* **Foundry resource**. Create a [Foundry resource](../../../../multi-service-resource.md) or see [Configure a Foundry resource](../../../concepts/configure-azure-resources.md). Alternatively, you can use a [Language resource](https://portal.azure.com/?Microsoft_Azure_PIMCommon=true#create/Microsoft.CognitiveServicesTextAnalytics).
* **A Foundry project**. For more information, see [Create a Foundry project](../../../../../ai-foundry/how-to/create-projects.md).
-
-### [Foundry (classic)](#tab/foundry-classic)
+### [Foundry (new)](#tab/foundry-new)
> [!NOTE]
-> This content refers to the [Foundry (classic)](https://ai.azure.com/) portal, which supports hub-based projects and other resource types. To confirm that you're using Foundry (classic), make sure the version toggle in the portal banner is in the **off** position. :::image type="icon" source="../../media/quickstarts/azure-ai-foundry/classic-foundry.png" border="false":::
-
+> This content refers to the [Foundry (new)](https://ai.azure.com/) portal, which supports only [Foundry projects](../../../../../ai-foundry/what-is-foundry.md) and provides streamlined access to models, agents, and tools. To confirm that you're using Foundry (new), make sure the version toggle in the portal banner is in the **on** position. :::image type="icon" source="../../media/quickstarts/azure-ai-foundry/new-foundry.png" border="false":::
-You can use [Foundry (classic)](https://ai.azure.com/) to:
+You can use [Foundry (new)](https://ai.azure.com/) to:
> [!div class="checklist"]
-> * Extract PII from conversations
-> * Extract PII from text
-> * Configure redaction policies
+>
+> * Extract personally identifiable information (PII) from text
+> * Configure redaction policies and excluded values
> * Review detected entities and confidence scores
+## Navigate to the Foundry (new) playground
-## Navigate to the Foundry (classic) playground
+The active project appears in the upper-left corner. To create a new project:
-1. In the left pane, select **Playgrounds**.
-1. Select the **Try Azure Language Playground** button.
+1. Open the project drop-down menu.
+1. Enter a project name or select an existing one.
+1. Select **Create project**.
-:::image type="content" source="../../media/quickstarts/azure-ai-foundry/foundry-playground-navigation.png" alt-text="Screenshot showing the Playgrounds navigation and the Try Azure Language Playground button in Foundry (classic)." lightbox="../../media/quickstarts/azure-ai-foundry/foundry-playground-navigation.png":::
+ :::image type="content" source="../../../media/new-foundry-homepage.png" alt-text="Screenshot of the Foundry (new) homepage.":::
-## Detect PII in the Foundry playground
+There are two ways to access the personally identifiable information (PII) interface:
-The **Language Playground** consists of four sections:
+1. Select the **Discover** tab from the upper right navigation bar to go to the **Models** page.
+ * In the search bar under models, enter **Azure** and press enter.
+ * Next, select **Azure-Language-Text-PII redaction** from the search results.
+ * Finally, select the **Open in Playground** button.
-| Section | Purpose |
-| --- | --- |
-| **Top banner** | Select the input language and choose a PII detection capability. |
-| **Left pane** | Set **Configuration** options such as API version, model version, and redaction policy. |
-| **Center pane** | Enter text for processing and review highlighted results. |
-| **Right pane** | View **Details** for each detected entity. |
+1. Select the **Build** tab from the upper right navigation bar.
+ * From the left navigation bar, select **Models**.
+ * Select the **AI services** tab.
+ * Next, select **Azure-Language-Text-PII redaction** to go to the playground.
-Select either **Extract PII from conversation** or **Extract PII from text** from the top banner tiles. Each capability targets a different scenario.
+## Extract PII from text
-### Extract PII from conversation
+The **Azure-Language-Text-PII redaction** model is used to identify and redact personally identifiable information in text. The interface provides configuration options to customize your detection and redaction preferences, as well as detailed output to review the detected entities and their confidence scores.
-**Extract PII from conversation** is designed to identify and mask personally identifiable information in conversational text.
+1. On the **Playground** tab, select **Azure Language—Text PII redaction** from the drop-down menu.
-In **Configuration** there are the following options:
+1. Select the sample text, use the paperclip icon to upload your text, or enter your own text.
-|Option |Description |
-|--------------------|-----------------------------------------|
-|Select API version | Select which version of the API to use. |
-|Select model version| Select which version of the model to use.|
-|Select text language| Select the language of your input text.|
-|Select types to include| Select the types of information you want to redact.|
-|Specify redaction policy| Select the method of redaction.|
-|Specify redaction character| Select the character used for redaction. Only available with the **CharacterMask** redaction policy.|
+1. In the **Configure** side panel, you can set the following options:
-After the operation completes, each detected entity is highlighted in the center pane with its type label displayed beneath it.
+ | Option | Description |
+ | --- | --- |
+ | **API version** | Select the API version that you prefer to use. |
+ | **Model version** | Select the model version that you prefer to use. |
+ | **Language** | Select the language in which your source text is written. |
+ | **Select types to include** | Select the PII types you want to redact. |
+ | **Value to exclude** | Specify values that you want to exclude from detection. For example, if you want to redact email addresses but want to exclude ones from a specific domain, you can enter that domain as an excluded value. |
+ | **Synonyms** | Provide alternative names for specific entity types. For example, if you select enter "Microsoft" as an excluded value, you can also specify specific synonyms such as `MSFT` and `Microsoft Corporation`. |
+ | **Policy type** | Choose the type of redaction policy to apply (character mask, entity mask, or no mask). |
-The **Details** section contains the following fields for each entity:
+1. After you make your selections, choose the **Detect** button. Detected entities are highlighted in the text and you can review the accompanying details in formatted text or as a JSON response.
-|Field | Description |
-|------|----------------------------|
-|Entity|The detected entity.|
-|Category| The entity type that was detected.|
-|Offset| The number of characters from the beginning of the line to the entity.|
-|Length| The character length of the entity.|
-|Confidence| The model's level of certainty that the entity type is correct.|
+ :::image type="content" source="../../media/quickstarts/azure-ai-foundry/new-foundry-text-detection.png" alt-text="Screenshot of PII text detection output in the Foundry playground." lightbox="../../media/quickstarts/azure-ai-foundry/new-foundry-text-detection.png":::
-:::image type="content" source="../../media/quickstarts/azure-ai-foundry/conversation-pii.png" alt-text="A screenshot showing detected PII entities highlighted in a conversation with entity details displayed in the right pane of the Foundry portal." lightbox="../../media/quickstarts/azure-ai-foundry/conversation-pii.png":::
+ | Field | Description |
+ | --- | --- |
+ | **Type** | The detected type. |
+ | **Confidence** | The model's level of certainty regarding whether it correctly identified an entity type. |
+ | **Offset** | The number of characters that the entity was detected from the beginning of the text. |
+ | **Length** | The character length of the entity. |
-Verify that each PII entity appears highlighted with the correct category label. If no entities appear, check that the input text contains recognizable PII patterns and that the **Types** filter includes the expected categories.
+Verify that the detected entities match the PII in your input text. You can use the **Edit** button to modify the **Configure** parameters and rerun detection as needed.
-### Extract PII from text
+## Extract PII from conversations 🆕
-**Extract PII from text** is designed to identify and mask personally identifiable information in text.
+The **extract PII from conversations** feature detects and masks personally identifying information within conversational text. This feature is designed to handle the unique structure and context of conversations, such as dialogues or chat logs.
-In **Configuration** you can select from the following options:
+1. On the **Playground** tab, select **Azure Language—Conversational PII redaction** from the drop-down menu.
-|Option |Description |
-|--------------------|-----------------------------------------|
-|Select API version | Select which version of the API to use. |
-|Select model version| Select which version of the model to use.|
-|Select text language| Select the language of your input text.|
-|Select types to include| Select the types of information you want to redact.|
-|Specify redaction policy| Select the method of redaction.|
-|Specify redaction character| Select the character used for redaction. Only available with the **CharacterMask** redaction policy.|
+1. Select a sample transcript, use the paperclip icon to upload your transcript, or input conversation turn text.
-After the operation completes, each detected entity is highlighted in the center pane with its type label displayed beneath it.
+1. Format your conversation with each turn on a new line and include speaker labels if possible. If participant ID, colon, message, or new line are not used consistently, the output may be unexpected. Here's an example of well-formatted conversation text:
-The **Details** section contains the following fields for each entity:
+ ```text
+ Speaker 1: Hello, how are you?
+ Speaker 2: I'm good, thank you! How about you?
+ Speaker 1: I'm doing well, thanks for asking.
+ ```
-|Field | Description |
-|------|----------------------------|
-|Entity|The detected entity.|
-|Category| The entity type that was detected.|
-|Offset| The number of characters from the beginning of the line to the entity.|
-|Length| The character length of the entity.|
-|Confidence| The model's level of certainty that the entity type is correct.|
-|Tags| The model's confidence scores for each identified entity subtype.|
+1. In the **Configure** side panel, set your preferred options for API version, model version, language, types, redaction policy, and excluded values.
-:::image type="content" source="../../media/quickstarts/azure-ai-foundry/text-pii.png" alt-text="A screenshot showing detected PII entities highlighted in text with entity details and confidence scores displayed in the right pane of the Foundry portal." lightbox="../../media/quickstarts/azure-ai-foundry/text-pii.png":::
+ | Option | Description |
+ | --- | --- |
+ | **API version** | Select the API version that you prefer to use. |
+ | **Model version** | Select the model version that you prefer to use. |
+ | **Language** | Select the language in which your source text is written. |
+ | **Select types to include** | Select the PII types you want to redact. |
+ | **Specify redaction character** | Choose the character used to mask sensitive text. |
-Verify that each PII entity appears highlighted with the correct category label. The **Tags** column shows subcategory confidence scores when applicable.
+1. After you make your selections, choose the **Detect** button. Detected entities are highlighted in the text and you can review the accompanying details in formatted text or as a JSON response.
-### [Foundry (new)](#tab/foundry-new)
+ :::image type="content" source="../../media/quickstarts/azure-ai-foundry/new-foundry-conversation-detection.png" alt-text="Screenshot of PII conversation detection output in the Foundry playground." lightbox="../../media/quickstarts/azure-ai-foundry/new-foundry-conversation-detection.png":::
-> [!NOTE]
-> This content refers to the [Foundry (new)](https://ai.azure.com/) portal, which supports only [Foundry projects](../../../../../ai-foundry/what-is-foundry.md) and provides streamlined access to models, agents, and tools. To confirm that you're using Foundry (new), make sure the version toggle in the portal banner is in the **on** position. :::image type="icon" source="../../media/quickstarts/azure-ai-foundry/new-foundry.png" border="false":::
+ | Field | Description |
+ | --- | --- |
+ | **Type** | The detected type. |
+ | **Confidence** | The model's level of certainty regarding whether it correctly identified an entity type. |
+ | **Offset** | The number of characters that the entity was detected from the beginning of the text. |
+ | **Length** | The character length of the entity. |
+Verify that the detected entities match the PII in your input conversation. You can use the **Edit** button to modify the **Configure** parameters and rerun detection as needed.
-You can use [Foundry (new)](https://ai.azure.com/) to:
+### [Foundry (classic)](#tab/foundry-classic)
+
+> [!NOTE]
+> This content refers to the [Foundry (classic)](https://ai.azure.com/) portal, which supports hub-based projects and other resource types. To confirm that you're using Foundry (classic), make sure the version toggle in the portal banner is in the **off** position. :::image type="icon" source="../../media/quickstarts/azure-ai-foundry/classic-foundry.png" border="false":::
+
+You can use [Foundry (classic)](https://ai.azure.com/) to:
> [!div class="checklist"]
+>
+> * Extract PII from conversations
> * Extract PII from text
-> * Configure redaction policies and excluded values
+> * Configure redaction policies
> * Review detected entities and confidence scores
+## Navigate to the Foundry (classic) playground
-## Navigate to the Foundry (new) playground
+1. In the left pane, select **Playgrounds**.
+1. Select the **Try Azure Language Playground** button.
-The active project appears in the upper-left corner. To create a new project:
+:::image type="content" source="../../media/quickstarts/azure-ai-foundry/foundry-playground-navigation.png" alt-text="Screenshot showing the Playgrounds navigation and the Try Azure Language Playground button in Foundry (classic)." lightbox="../../media/quickstarts/azure-ai-foundry/foundry-playground-navigation.png":::
-1. Open the project drop-down menu.
-1. Enter a project name or select an existing one.
-1. Select **Create project**.
+## Detect PII in the Foundry playground
- :::image type="content" source="../../../media/new-foundry-homepage.png" alt-text="Screenshot of the Foundry (new) homepage":::
+The **Language Playground** consists of four sections:
+| Section | Purpose |
+| --- | --- |
+| **Top banner** | Select the input language and choose a PII detection capability. |
+| **Left pane** | Set **Configuration** options such as API version, model version, and redaction policy. |
+| **Center pane** | Enter text for processing and review highlighted results. |
+| **Right pane** | View **Details** for each detected entity. |
-There are two ways to access the PII interface:
+Select either **Extract PII from conversation** or **Extract PII from text** from the top banner tiles. Each capability targets a different scenario.
-1. Select the **Discover** tab from the upper right navigation bar to go to the **Models** page.
- * In the search bar under models, enter **Azure** and press enter.
- * Next, select **Azure-Language-Text-PII redaction** from the search results.
- * Finally, select the **Open in Playground** button.
+### Extract PII text
-1. Select the **Build** tab from the upper right navigation bar.
- * From the left navigation bar, select **Models**.
- * Select the **AI services** tab.
- * Next, select **Azure-Language-Text-PII redaction** to go to the playground.
+**Extract PII from text** is designed to identify and mask personally identifying information in text.
+
+In **Configuration** you can select from the following options:
+
+| Option | Description |
+| --- | --- |
+| Select API version | Select which version of the API to use. |
+| Select model version | Select which version of the model to use. |
+| Select text language | Select the language of your input text. |
+| Select types to include | Select the types of information you want to redact. |
+| Specify redaction policy | Select the method of redaction. |
+| Specify redaction character | Select the character used for redaction. Only available with the **CharacterMask** redaction policy. |
+
+After the operation completes, each detected entity is highlighted in the center pane with its type label displayed beneath it.
+The **Details** section contains the following fields for each entity:
-## Detect PII in the Foundry playground
+| Field | Description |
+| --- | --- |
+| Entity | The detected entity. |
+| Category | The entity type that was detected. |
+| Offset | The number of characters from the beginning of the line to the entity. |
+| Length | The character length of the entity. |
+| Confidence | The model's level of certainty that the entity type is correct. |
+| Tags | The model's confidence scores for each identified entity subtype. |
-The **extract PII from text** feature detects and masks personally identifiable information within written content.
+:::image type="content" source="../../media/quickstarts/azure-ai-foundry/text-pii.png" alt-text="A screenshot showing detected PII entities in text displayed in the Foundry portal." lightbox="../../media/quickstarts/azure-ai-foundry/text-pii.png":::
-1. On the **Playground** tab, select the sample text, use the paperclip icon to upload your text, or enter your own text.
+Verify that each PII entity appears highlighted with the correct category label. The **Tags** column shows subcategory confidence scores when applicable.
+
+### Extract PII conversations
-1. Select the **Configure** button. In the **Configure** side panel, set the following options:
+**Extract PII from conversation** is designed to identify and mask personally identifying information in conversational text.
+
+In **Configuration** there are the following options:
| Option | Description |
-|--|--|
-| **API version** | Select the API version that you prefer to use. |
-| **Model version** | Select the model version that you prefer to use. |
-| **Language** | Select the language in which your source text is written. |
-| **Types** | Select the types of information you want to redact. |
-| **Specify redaction policy** | Select the method of redaction. |
-| **Excluded values** | Select the values that you want to exclude. |
-| **Synonyms** | Select a category for your redaction type values to target related synonyms. |
+| --- | --- |
+| Select API version | Select which version of the API to use. |
+| Select model version | Select which version of the model to use. |
+| Select text language | Select the language of your input text. |
+| Select types to include | Select the types of information you want to redact. |
+| Specify redaction policy | Select the method of redaction. |
+| Specify redaction character | Select the character used for redaction. Only available with the **CharacterMask** redaction policy. |
-After you make your selections, choose the **Detect** button. Detected entities are highlighted in the text and you can review the accompanying details in formatted text or as a JSON response:
+After the operation completes, each detected entity is highlighted in the center pane with its type label displayed beneath it.
-> [!NOTE]
-> The Foundry (new) playground currently supports **text PII** detection. For **conversation PII**, use the **Foundry (classic)** tab.
+The **Details** section contains the following fields for each entity:
| Field | Description |
-|--|--|
-| **Type** | The detected type. |
-| **Confidence** | The model's level of certainty regarding whether it correctly identified an entity type. |
-| **Offset** | The number of characters that the entity was detected from the beginning of the text. |
-| **Length** | The character length of the entity. |
+| --- | --- |
+| Entity | The detected entity. |
+| Category | The entity type that was detected. |
+| Offset | The number of characters from the beginning of the line to the entity. |
+| Length | The character length of the entity. |
+| Confidence | The model's level of certainty that the entity type is correct. |
-Verify that the detected entities match the PII in your input text. You can use the **Edit** button to modify the **Configure** parameters and rerun detection as needed.
\ No newline at end of file
+:::image type="content" source="../../media/quickstarts/azure-ai-foundry/conversation-pii.png" alt-text="A screenshot showing detected PII entities highlighted in a conversation with entity details displayed in the right pane of the Foundry portal." lightbox="../../media/quickstarts/azure-ai-foundry/conversation-pii.png":::
+
+Verify that each PII entity appears highlighted with the correct category label. If no entities appear, check that the input text contains recognizable PII patterns and that the **Types** filter includes the expected categories.
+
+
+
+---