Microsoft Combats Deceitful Deepfakes With AI Unmasking Edits Unseen By Humans
By Thanussha Priyah, 03 Sep 2020
Image via Shutterstock
Deepfake technology is becoming a growing issue regarding deception and misinformation in the digital age. So, Microsoft has stepped up with two new authentication tools to confront mishandled content online.
The first tool is called the Microsoft Video Authenticator, which studies images and videos to gauge “a percentage chance, or confidence score, that the media is artificially manipulated.”
It functions by detecting the “blending boundary of the deepfake and subtle fading or greyscale elements,” which may not be obvious to the human eye.
The second feature is part of Microsoft’s cloud service Azure, and it lets creators label their images and videos with “digital hashes and certificates.” These will then appear in the media’s metadata as it gets shared online through various mediums.
The tags can be matched through a browser extension tool, allowing internet users to find out about the origin of the content, and whether it has been altered since.
Microsoft hopes to arm discerning internet users with the ability to differentiate real content from synthetic media through these new technologies.
The company also noted that combative AI tools have to keep upgrading constantly to match deepfake technology’s rapid evolvement.
How does the Video Authenticator tool work?— Microsoft On the Issues (@MSFTIssues) September 1, 2020
It detects the blending boundary of a #deepfake and subtle fading or greyscale elements that might not be detectable by the human eye. 👁️ Learn more: https://t.co/ovruz23BoQ (2/2) pic.twitter.com/sLmcnrQAcr
[via Mashable, cover image via Shutterstock]
More related news