Facebook AI Software to Detect the Origins of Deepfake Images

Facebook researchers have developed artificial intelligence software to identify so-called "deepfakes" and trace their origins using reverse engineering generative models.

Facebook AI Software to Detect the Origins of Deepfake Images

Facebook Scientists Develop AI Software to Detect Deepfakes and Trace Their Origins

Facebook Scientists of the artificial intelligence department revealed important progress in the field of fighting fake images and information and said that they have developed a new artificial intelligence software, which can not only detect and identify deepfake images but also be able to identify and know their creators by using reverse engineering.

Facebook researchers claimed that they have created an artificial intelligence program that reverses deepfakes to see how they were made and where they originated. This software can make it easier to detect and track deepfakes in real-world settings. The deepfake image itself is often the only information detectors you have to work with.

Facebook's AI software works on deepfakes through its network to look for defects left during the image production process, which the researchers say alters the digital image's fingerprint. 

In digital photography, fingerprints are used to identify and point out the digital camera used to produce an image. These fingerprints are also unique patterns that can equally be used to detect and trace the generative model from which the image came.

Facebook scientists noted that the new AI software will give researchers and practitioners tools to better investigate deepfake images and misinformation using reverse engineering generative models, as well as open up new directions for future research.

The Scientific World

The Scientific World is a Scientific and Technical Information Network that provides readers with informative & educational blogs and articles. Site Admin: Mahtab Alam Quddusi - Blogger, writer and digital publisher.

Previous Post Next Post