Export and delete your customer data in Language Understanding (LUIS) in Azure AI services

Important

LUIS will be retired on October 1st 2025 and starting April 1st 2023 you will not be able to create new LUIS resources. We recommend migrating your LUIS applications to conversational language understanding to benefit from continued product support and multilingual capabilities.

Delete customer data to ensure privacy and compliance.

Summary of customer data request features​

Language Understanding Intelligent Service (LUIS) preserves customer content to operate the service, but the LUIS user has full control over viewing, exporting, and deleting their data. This can be done through the LUIS web Azure portal or the LUIS Authoring (also known as Programmatic) APIs.

Note

This article provides steps about how to delete personal data from the device or service and can be used to support your obligations under the GDPR. For general information about GDPR, see the GDPR section of the Microsoft Trust Center and the GDPR section of the Service Trust portal.

Customer content is stored encrypted in Microsoft regional Azure storage and includes:

  • User account content collected at registration
  • Training data required to build the models
  • Logged user queries used by active learning to help improve the model
    • Users can turn off query logging by appending &log=false to the request, details here

Deleting customer data

LUIS users have full control to delete any user content, either through the LUIS web portal or the LUIS Authoring (also known as Programmatic) APIs. The following table displays links assisting with both:

User Account Application Example Utterance(s) End-user queries
Portal Link Link Link Active learning utterances
Logged Utterances
APIs Link Link Link Link

Exporting customer data

LUIS users have full control to view the data on the portal, however it must be exported through the LUIS Authoring (also known as Programmatic) APIs. The following table displays links assisting with data exports via the LUIS Authoring (also known as Programmatic) APIs:

User Account Application Utterance(s) End-user queries
APIs Link Link Link Link

Location of active learning

To enable active learning, users' logged utterances, received at the published LUIS endpoints, are stored in the following Azure geographies:

With the exception of active learning data (detailed below), LUIS follows the data storage practices for regional services.

Note

As of January 20th 2021, The luis.au and luis.eu portals have been consolidated into a single LUIS portal. If you were using one of these portals, instead access LUIS at luis.ai.

Europe

Europe Authoring (also known as Programmatic APIs) resources are hosted in Azure's Europe geography, and support deployment of endpoints to the following Azure geographies:

  • Europe
  • France
  • United Kingdom

When deploying to these Azure geographies, the utterances received by the endpoint from end users of your app will be stored in Azure's Europe geography for active learning.

Australia

Australia Authoring (also known as Programmatic APIs) resources are hosted in Azure's Australia geography, and support deployment of endpoints to the following Azure geographies:

  • Australia

When deploying to these Azure geographies, the utterances received by the endpoint from end users of your app will be stored in Azure's Australia geography for active learning.

United States

United States Authoring (also known as Programmatic APIs) resources are hosted in Azure's United States geography, and support deployment of endpoints to the following Azure geographies:

  • Azure geographies not supported by the Europe or Australia authoring regions

When deploying to these Azure geographies, the utterances received by the endpoint from end users of your app will be stored in Azure's United States geography for active learning.

Switzerland North

Switzerland North Authoring (also known as Programmatic APIs) resources are hosted in Azure's Switzerland geography, and support deployment of endpoints to the following Azure geographies:

  • Switzerland

When deploying to these Azure geographies, the utterances received by the endpoint from end users of your app will be stored in Azure's Switzerland geography for active learning.

Disable active learning

To disable active learning, see Disable active learning.

Next steps