For instance, the notion of seeing Nicholas Cage’s likeness appearing in a number of films is inherently funny. In humorous uses of deepfakes, the absurdity of the artifice is generally the source of the humor. In general, a few uses come to mind when deepfakes are talked about today: parody, politics, and pornography. These media have been steadily rising in both their infamy and popularity in recent years on learning what they are, one could imagine a number of purposes for which deepfakes could be used, some nefarious, others harmless.
As such, most often, source images are of those who can easily be found online: celebrities and politicians.ĬW: Rape and Sexual Assault, Sexual Abuse. A machine learning algorithm uses this database to create a video or image that, through an immense amount of data points, looks believable, moving as if a human would move. In other words, a large database of source images and videos must be made available in order for deepfakes to appear realistic. In order to create a highly believable final manipulated image the technology must be able to determine the data points to match the image to the final manipulated one. It allows users to superimpose existing images and videos onto source images or videos using a machine learning technique or programme. 1 Its name derives from a Reddit user of the same name, 2 though it has also been said to be a portmanteau of “deep-learning” and “fake.” 3 Tyrone Kirchengast explains the mechanism behind deepfake technology as follows: Issues of Consent, Representation, and Exploitation in Deepfake Pornographyĭeepfakes are a kind of synthetic media that use machine learning to create an artificially manipulated image or video with someone’s likeness or in some cases, an entirely artificially generated person.