Frequently asked questions about Copilot in Word

Copilot in Word 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 Word is continuously evaluated with user online feedback. 

Many languages are supported in Copilot scenarios. The quality is expected to be highest in English, while in other languages the quality is expected to be improved over time. More languages and locales will be added in the future. Find more information at Microsoft Copilot supported languages.

The suggestions from Copilot in Word are from AI, and we highly recommend users to review the suggestions before accepting them. 

Also, Copilot is currently limited in the number of words it can process per prompt. For more information about processing longer files, see Keeping it short and sweet: a guide on the length of documents that you provide to Copilot.

Receiving coaching on your writing is currently in a preview release state. As such, the "Get coaching" option is limited to a smaller set of supported languages and regions while we evaluate quality of output and ensure that it meets our high quality and experience bar. To learn more, see Use Coaching to review content in Word for the web.​​​​​​​

Copilot in Word has been reviewed by our Responsible AI (RAI) team. We follow RAI principles and have implemented: 

• Responsible AI handling pipeline to mitigate the risks like harmful, inappropriate content.

• In product user feedback with which users can report offensive content back to Microsoft.

Copilot in Word includes filters to block offensive language in the prompts and to avoid synthesizing suggestions in sensitive contexts. We continue to work on improving the filter system to more intelligently detect and remove offensive outputs. If you see offensive outputs, please submit feedback in Word using the thumbs up/thumbs down feedback in the Copilot UI so that we can improve our safeguards. Microsoft takes this challenge very seriously, and we are committed to addressing it.

Copilot in Word writes in a fluent, grammatically correct way, but the content it generates can be inaccurate or inappropriate. It can’t understand meaning or evaluate accuracy, so be sure to read over what it writes, and use your judgment.  

While these features work 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.

As with any AI generated content it's a great tool to get started, but it's important that you review, edit, and verify anything it creates for you.

Copilot in Word generates content based on language patterns it has found throughout the internet. Sometimes its results will be very similar to existing internet content. Or, Copilot might generate the same or very similar content for multiple people who are prompting Copilot in the same way. 

For example, everyone who instructs Copilot to create a "recipe for apple pie" is likely to wind up with identical or nearly identical content. When originality is a concern, check for similarity to online sources (on the Home tab, go to Editor, and scroll down to Similarity). 

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 at home as part of Copilot Pro (with your personal Microsoft account), see Copilot Pro: Microsoft 365 apps and your privacy.