Well, what convenient timing... "Deepfake detectors can be defeated, computer scientists show for the first time"
(www.eurekalert.org)
BREAKING NEWS
You're viewing a single comment thread. View all comments, or full comment thread.
Comments (57)
sorted by:
As a researcher (and experts) in this area - Good luck.
The problem with any deep fake detector:
The model that generates deep fakes are going to directly be tested against the detector. That enables the deep fake generators to improve.
It’s a classic arms race problem. If you make the detector public, you can expect the deep fake models to incorporate that input very quickly.