Microsoft drafts new AI guidelines and stops facial recognition in Azure
Microsoft will set tighter guidelines for how it builds artificial intelligence systems. The Responsible AI Standard contains rules for voice and facial recognition. Microsoft will stop with the latter in part, because there are too many risks involved.
Microsoft say that it is the most recent version of Responsible AI Standard has released. An earlier version of it was only available internally. That guideline states what Microsoft does and cannot do with artificial intelligence and the tools it makes for it. Those guidelines set out how Microsoft handles issues such as privacy, security, and reliability, as well as inclusiveness and transparency. The tools that Microsoft builds that use artificial intelligence, such as in Azure or Windows, must meet all those standards.
In addition to the guideline, the company also shares examples of where it has fallen short in previous years. For example, the company cites a study that showed that the speech-to-text technology it built failed almost twice as often among black users compared to white ones. Microsoft also gives an example of how the company built additional layers of protection into Azure’s Custom Neural Voice when it emerged that there was a chance it could be used to impersonate other people.
In the future, Microsoft wants to make it less easy to deploy AI tools that can be misused for such things. One of the most notable decisions Microsoft has made in response to the guidelines is that the company makes some facial recognition technologies more difficult in Azure to use. Users who want to use the Azure Face API, Computer Vision, or Video Indexer must request access specifically. Existing users have one year to do so. In that case, they must indicate what they use the service for, provide examples of this and make plans to prevent abuse. Microsoft can revoke access in case of abuse.
It is not about facial recognition tools in general, but specifically about tools that can read emotions from facial recognition images. Finding out someone’s gender, age, smile, facial hair or make-up on the basis of images is also restricted. New users will no longer be able to access it at all. Existing users have one year to phase out use. Standard applications, including recognizing, for example, glasses in photos or removing noise and lighting in an image, will remain available.