Frequently asked questions about Copilot in Loop

Copilot combines the power of large language models (LLMs) with Loop to help you more easily ask questions, create content, and collaborate with your peers.

We recently removed some Copilot integrations from the Loop app in preparation to synchronize the experience with Microsoft 365 Copilot Pages experience. The differences were leading to user confusion. 

We're actively working to bring some Copilot experiences from Microsoft 365 Copilot Pages into Loop and upgrade the Copilot side pane to match the same model used in the chat + pages experience. This will ensure responses and interactions are consistent across products. IN the meantime, you can still access the older  Copilot integration in Loop by selecting the  Copilot button in the upper-right corner to refine content. We'll share progress through the Message Center and appreciate your patience as we make these improvements.

A Microsoft 365 Copilot license is required to use Copilot in Loop. In addition, a Microsoft 365 E3 or E5 license is required to purchase a Microsoft 365 Copilot license. Learn more about purchasing Microsoft 365 Copilot.

People with a Microsoft 365 Copilot license can use Copilot in Loop and share with others within their organizations. People in their organizations without a Copilot license have a limited ability to view and edit content created with Copilot. The following table details what people with and without Copilot licenses can do with Copilot.

Copilot in Loop was evaluated through extensive manual and automatic testing on top of Microsoft internal usage and public data. More evaluation was performed over custom datasets for offensive and malicious prompts (user questions) and responses. In addition, Copilot in Loop is constantly evaluated with user online feedback (thumbs up and down feedback function).

Currently, Copilot in Loop is only aware of web or enterprise search results and cannot be used to ask about the Loop Workspace, or to make updates to the currently-open Page.

Many languages are supported in Copilot in Loop, and more languages will be supported in the future. To learn more, see Supported languages for Microsoft Copilot.

Copilot in Loop cannot be customized. Admins can choose to enable Copilot for specific users and users can choose what questions to ask Copilot based on their needs.

Copilot in Loop will best respond when users do the following:

1. Limit questions to topics that can be found in web or enterprise searches.

2. Use supported languages. Copilot will respond in supported languages.

3. Ensure to provide enough context in your asks to Copilot.

Copilot in Loop supports many languages. For more information, see Microsoft Copilot supported languages.

Copilot and Microsoft 365 are built on Microsoft's comprehensive approach to security, compliance, and privacy.

For more information about privacy, see the following information:

• If you're using Microsoft 365 Copilot in your organization (with your work or school account), see Data, Privacy, and Security for Microsoft 365 Copilot.

• If you're using Copilot in Microsoft 365 apps for home (with your personal Microsoft account), see Copilot in Microsoft 365 apps for home: your data and privacy.

Generative AI features strive to provide accurate and informative responses, based on the data available. However, answers may not always be accurate as they are generated based on patterns and probabilities in language data. Use your own judgment and double check the facts before making decisions or taking action based on the responses.

While these features have mitigations in place to avoid sharing unexpected offensive content in results and take steps to prevent displaying potentially harmful topics, you may still see unexpected results. We're constantly working to improve our technology to proactively address issues in line with our responsible AI principles.

If you find a generated response that is incorrect or if you encounter harmful or inappropriate content, please provide that feedback by clicking thumbs down in the summary and providing additional comments in the feedback form. This feedback helps us to improve and minimize this content in the future.