After an anonymous Reddit user going by the name “deepfakes” created a forum in 2017 dedicated to making pornographic videos of female celebrities using deep learning technology, experts warned that digitally manipulated audio-visual material was a serious threat with wide-ranging consequences. Since then, the topic of deepfakes has received considerable media coverage, academic discussion, and calls for legislative responses.
A handful of states have passed legislation making some deepfakes illegal, and several tech platforms have announced policies against digital misinformation, but there is no clear, comprehensive, and effective federal response to the problem. Indeed, there is not even broad agreement that deepfakes are a serious problem in need of response. In the view of some individuals and organizations, concerns about deepfakes are overblown, and proposed policy responses risk endangering free speech.
Advertisement
Those who argue that deepfakes are not a serious problem point out that the much-feared wave of deepfakes involving presidential candidates or other political figures has so far failed to materialize. It is true that relatively few political deepfakes have gained traction in the United States. But complacency about deepfakes is both sexist and shortsighted, and demonstrates a fundamental lack of understanding of the threat that they pose to autonomy, truth, and democracy.
According to Sensity (formerly Deeptrace Lab), a company that tracks synthetic media threats, there were nearly 15,000 deepfake videos online in 2019, an 84 percent increase from the year before. Of those, 96 percent were pornographic, and all of of these pornographic videos, which have received more than 134 million views, featured women. Sites using the widely publicized source code to make pornographic deepfakes have flourished since 2018, including websites and applications offering to create custom deepfakes of former girlfriends, neighbors, and strangers based on innocuous photos and videos. Like all forms of nonconsensual pornography, deepfake pornography violates victims' autonomy, injures their reputations and careers, and jeopardizes their physical, psychological, and emotional welfare.
Advertisement
To characterize deepfakes as a minor issue is to dismiss the sexual exploitation of women and girls as unimportant. In addition, while women are often the first targets of technological abuse, they are rarely the last. A society that ignores or downplays a technological threat so long as it is used to “only” sexually exploit women will be woefully unprepared to respond when that technology is weaponized against other people and for other causes. Deepfakes will increasingly be used against the powerful as well as the vulnerable. We already are witnessing the use of deepfakes to target political figures and to sow civil unrest.
In addition to ignoring or trivializing the harm of deepfakes because they are not the targets, those who object to regulating deepfakes often claim that any cure will be worse than the disease. According to the marketplace of ideas theory of free speech, “the best test of truth is the power of the thought to get itself accepted in the competition of the market,” in Justice Oliver Wendell Holmes’s words. In this view, the market will see to it that deepfakes, by virtue of their falsity, will be eventually abandoned as inferior products in favor of the truth.
But especially given the state of disinformation in America, such a belief can most charitably be described as willfully ignorant. Lies, especially those that serve the interests of those in power, have always had a competitive advantage over the truth, and truthful speech frequently gets drowned out by fake, misleading, and salacious content.
Advertisement
Civil libertarians are fond of saying that the best cure for bad speech is more speech, or, in Justice Louis Brandeis’s words: “If there be time to expose through discussion the falsehood and fallacies, to avert the evil by the processes of education, the remedy to be applied is more speech, not enforced silence.” But deepfakes are particularly resistant to the remedy of counter-speech precisely because their impact is so immediate and intractable — there is no time to avert their harm before it’s done, and education after the fact will either have limited effect or even backfire. Repeated exposure to false information, even when presented for the purposes of correction, increases the likelihood that false information will be remembered as true.
Deepfakes are a form of unanswerable speech. Like the unauthorized publication of a person’s actual nude image, the dissemination of a home address, or the disclosure of one’s sexual orientation, deepfakes are not ideas that can simply be countered with different and better ideas.
Deepfakes are dangerous not only because they make people believe that false things are true, but also because they make people believe that true things are false. There is as much danger in a video falsely portraying a person not engaging in misconduct as there is in a video falsely portraying a person engaging in misconduct, especially if that person is an elected official seeking to cover up his misdeeds. Perhaps most fundamentally, by creating doubt about what is true and what is false, deepfakes undermine both the capacity and the desire to separate truth from falsity. In an era of “alternative facts” and mass conspiracy theories, such confusion will only worsen our descent into tribalism and violence, leaving our society ever more vulnerable to the manipulations of power.
Advertisement
As the philosopher Hannah Arendt wrote in 1974, “If everybody always lies to you, the consequence is not that you believe the lies, but rather that nobody believes anything any longer. …A people that no longer can believe anything cannot make up its mind. It is deprived not only of the capacity to act but also of its capacity to think and to judge. And with such a people you can then do what you please.” There is a far darker future ahead if we do not heed her warning.
Mary Anne Franks is a professor of law at the University of Miami School of Law. She is author of “The Cult of the Constitution.”
