New AI Tech Blinds Computer Facial Recognition Systems

Jan 29, 2018

Facial recognition can be as scary as it is useful, but Tel Aviv startup D-ID aims to make it considerably harder for unauthorized others to use our personal photos for facial recognition matching.

D-ID’s founders Eliran Kuta, Gil Perry and Sella Blondheim – (the founders from left to right)PRNEWSFOTO/D-ID

Computers can scan through thousands of images much faster and more accurately than a human, making automated facial matching systems invaluable for speeding up security checks at border controls or for more mundane tasks such as unlocking your smartphone.

This is of obvious concern to privacy advocates, but we could all benefit from a little more security when it comes to our identity, especially in an age when digital images of our faces are so often and so widely circulated.

If your face is used instead of a password, then someone able to replicate your facial image may be able to gain unauthorized access and, once compromised, a password is much easier to change than your face.

What D-ID claims to do is to distort photographed faces in such a way that leaves them looking unaltered to the human eye yet unrecognizable to facial recognition algorithms.