Does Microsoft have the rights to all data used to train the azure translation service

Question

Hello,

I refer to this article https://www.eurogamer.net/valve-says-ai-generated-content-policy-goal-is-not-to-discourage-the-use-of-it-on-steam with this quote

"Our priority, as always, is to try to ship as many of the titles we receive as we can," Valve said, noting the introduction of AI may make this process harder as it is not always easy to know when a developer has "sufficient rights in using AI to create assets, including images, text, and music". Valve then touched on the legal uncertainty around the use of certain AI generated assets, stating "it is the developer's responsibility to make sure they have the appropriate rights to ship their game".

With this in mind, is it written anywhere publicly that Microsoft has the rights required to use the training data used to train the translation service? If not written publicly is it correct to assume they do have the rights to use the training data>

Answer 1

@Chris Aeberhard Azure translator service does not use the data for any purpose. Please see this page from translator service for compliance

Azure Cognitive Services Translator is a cloud-based machine translation service and is part of the Azure Cognitive Services family of cognitive APIs for building intelligent apps. Customer data submitted for translation to Azure Cognitive Services Translator (both standard and custom models), Speech service, the Microsoft Translator Speech API, and the text translation features in Microsoft Office products are not written to persistent storage. There will be no record of the submitted text or voice, or any portion thereof, in any Microsoft data center. The audio and text will not be used for training purposes either.

If this answers your query, do click Accept Answer and Yes for was this answer helpful. And, if you have any further query do let us know.