Deepfakes: The Lawless New World of Involuntary Porn

0 Comment
Deepfakes and Pornography

The Internet has been abuzz about “deepfakes” in the last couple of months… And for good reason.  “Deepfakes” is the term coined to describe images and videos, generated by artificial intelligence, where a person’s head is superimposed onto another person’s body in a pornographic scene (basically, face-swapping).  This enables creation of what appears to be actual porn involving a non-consenting person.  The problem with deepfakes is that they are extremely realistic with the stitching of images being almost undetectable in many cases.  Therefore, it is very difficult to discern whether the video is in fact fake or not.  This can put celebrities and private citizens in a predicament when these doctored videos are shared on the Internet.  There is much to be discussed in terms of legalities, ethics, and the future.

Deepfakes, i.e. involuntary porn, originated from a Reddit user who went by the name, “deepfakes.”  Using software, deepfakes would perform the face swap and then post to a Reddit forum.  Easy-to-use applications sprouted up with the ability to create these types of composite images/videos.  Other people then started posting too.  Deepfakes of Taylor Swift, Emma Watson, and other celebrities appeared on the Reddit forum.  Over just a few months, the forum gained thousands of users, and in so doing, sparked a conversation about ethics, porn, and AI.

The law typically has to catch up with technology, and not the other way around.  This is because technology moves much faster than government!  As deepfakes appeared on the scene relatively recently, there isn’t really a clear cut answer at this time as to the legal questions surrounding these doctored videos.  The operators of Reddit, YouTube, Pornhub, and Twitter have not waited for legislation, and instead have taken matters into their own hands.  They each modified their Terms of Use to prohibit deepfakes from being shared to their sites.  As non-governmental, privately-owned companies, they have every right to do this….

The U.S. government banning the making and sharing of deepfakes (made without consent) is another story.  In the United States, the First Amendment to the Bill of Rights provides:

Congress shall make no law … abridging the freedom of speech…

Creating a deepfake could be viewed as a form of speech – a comment, parody/satire, or other item of protected speech.  A government law prohibiting it could be considered unconstitutional censorship.  The First Amendment is unique to the U.S., and other countries do not necessarily guarantee such freedom of speech to their citizens.  It is therefore possible that those other countries could ban deepfakes under their own laws.

Along First Amendment lines, deepfakes may be argued to be a “fair use” of copyrighted material.  Copyright provides protection to authors of creative works, like photos and films.  Federal law enables an author to prevent copying or use of his/her copyrighted work without permission.  However, an exception to right is fair use.  The doctrine of fair use allows the copying or use of portions of another’s copyrighted work for certain purposes, such as comment, criticism, or parody.  This exception, closely related to the First Amendment, prevents the copyright of one person from impinging on the free speech of others.  Deepfakes can be argued to be a fair use-defensible parody or satire.  That doesn’t mean though that the argument would carry the day.  Until someone who has been “deepfaked” has their day in court to create a precedent, this is all just speculation.

While federal copyright law allows authors to control the copying and dissemination of their creative works of authorship, state laws providing the right of publicity give people the right to control the use of their names and likenesses (images).  The caveat for this right of publicity is that the control is for the use of their images for commercial purposes.  Accordingly, if a deepfake is not used in advertising or trade, it likely does not violate a person’s right to publicity.

Also provided by state laws is the protection against defamation, i.e. damaging false statements being disseminated about someone.  Libel, a form of defamation, is the act of making false statements about another that damages his/her reputation and is printed or broadcast over the media.  Unlike for private citizens, public figures have the extra burden to show that the defamation was made with malicious intent and was not just fair comment.  Claiming a deepfake is defamatory is a possible avenue for fighting the practice since the person portrayed in the video was not actually in it, and did not perform the acts depicted.  In theory, a plaintiff could win a defamation claim against a person who posts a deepfake without his/her consent, but such cases are expensive, difficult to win, and almost impossible to prosecute when the publishers are anonymous or outside of U.S. borders.  Perhaps someone will be up for the challenge though at some point.

Unlike the FALSE nature of a defamatory remark, revenge porn, as we have typically known it, is the posting of sexually-explicit REAL images of another without their consent.  Prior to the recent emergence of deepfake technology, a person could avoid having realistic sexy images of his/herself posted to the Internet by refraining from taking or sharing such.  Now, keeping sexy images private is no safety-net for one’s online reputation, as hyper-realistic substitute images can be produced.  Currently, in the U.S., there is no federal revenge porn prohibition, but 38 states and Washington D.C. have enacted bans.  Each law is written with slightly different language – Some make it illegal to share an “intimate image,” “nude photos,” “sexual photos,” “sexually-explicit photos,” or the like.  This begs the question of when the nude body is not that of the person depicted (in the face), can it be defined as a sexually-explicit photo of the facially depicted person for the purpose of revenge porn?  This has yet to be tried and determined.

For now, the intersection of deepfakes and the law leaves us with more questions than answers.  Over time, things will likely become clearer.  Congress will act by passing legislation, and/or court decisions will shed some light on the issue.  In the meantime, this is just one of the many developing legal and ethical issues being ushered in by the advancement of artificial intelligence.  It’s a new world, folks… It’s a new world.

Share This:

0 Comments

Leave a Comment