Advances in artificial intelligence and deep neural networks have led to a rise in synthetic media, i.e., automatically and artificially generated or manipulated photo, audio, and video content. Synthetic media today is highly believable and “true to life”; so much so that we will no longer be able to trust what we see or hear is unadulterated and genuine. Among the different forms of synthetic media, the most concerning forms are deepfakes and general adversarial networks (GANs). For IT professionals, it is important to understand what these new phenomena are. In this article, we explain what deepfakes and GANs are, how they work and discuss the threats and opportunities resulting from these forms of synthetic media.
[1]
Tim C. Kietzmann,et al.
Deepfakes: Trick or treat?
,
2020,
Business Horizons.
[2]
拓海 杉山,et al.
“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告
,
2017
.
[3]
Luc Van Gool,et al.
Pose Guided Person Image Generation
,
2017,
NIPS.
[4]
Yoshua Bengio,et al.
Generative Adversarial Nets
,
2014,
NIPS.
[5]
Jeff Donahue,et al.
Large Scale GAN Training for High Fidelity Natural Image Synthesis
,
2018,
ICLR.
[6]
Robert Chesney,et al.
Deepfakes and the New Disinformation War
,
2018
.