In an era of widespread political disinformation deployed by bad actors to influence democracies, Facebook has faced criticism for its policy of allowing advertisers to use its platform to spread false messages, particularly about politics. Facebook CEO Mark Zuckerberg has argued that such a policy is necessary to protect free speech, and that he doesn’t see Facebook’s role as one of censoring political messages. Certainly, the question of how to handle the spread of disinformation on social media networks is a tricky one, particularly during a time when the president’s reelection campaign overtly makes false claims in social media ads, most notably on Facebook, to influence the voting public.
Rival social network Twitter has decided to address the problem by banning political ads on Twitter altogether, neatly sidestepping the issue by refusing to participate in it in any capacity. But even in the face of ongoing, intense criticism and action taken on other platforms, Zuckerberg has remained steadfast in his opinion, positioning Facebook as a platform that promotes free speech instead of one that polices the political views of its users. However, amid the intensity of the criticism directed at the social networking giant, Facebook has recently announced it would ban “deepfakes” on the site, an apparent concession to those who are worried about social media’s role in facilitating the spread of false information.
So-called “deepfakes” are the result of new technology, made possible by advances in machine learning using neural networks, that can appear to show evidence of a person saying or doing something that they did not in fact say or do by manipulating video to superimpose an image of one person’s face onto another person’s head with near-perfect accuracy. Such videos can be difficult or impossible to detect, even by experts, and as the technology advances deepfakes become even more convincing and easy to make.
While these moves are certainly steps in the right direction, they are likely not enough to stop the spread of fake news
Researchers and political observers around the world have understandably voiced concerns about the potential impact of deepfakes on the spread of information, as the very existence of deepfakes causes one to call into question the legitimacy of videos depicting well-known political figures, which were once considered ironclad evidence of a person’s speech and conduct. To illustrate this point, director Jordan Peele created a deepfake that appears to depict President Obama delivering a warning about the spread of disinformation. Even more disturbingly, deepfakes have also been used to create pornographic videos appearing to depict various well-known celebrities, in violation of these celebrities’ rights to control how their images are used in public forums.
As the dangers that deepfakes pose to individuals and to society as a whole are clear, it’s no surprise that Facebook has taken the step of banning this type of video on its site. However, given the extent of the spread of misinformation online, this action alone is not nearly enough to ensure that bad actors cannot subvert democracies by spreading fake news. For one, deepfakes are difficult to detect, even using computer analysis; as such, Facebook launched the Deep Fake Detection Challenge in an attempt to improve the technology that can determine whether a video has been digitally manipulated. And while Facebook continues to allow the spread of falsehoods in the form of political advertisements, the company has also partnered with independent fact-checkers with the aim of informing users when they are encountering false information.
While these moves are certainly steps in the right direction, they are likely not enough to stop the spread of fake news, especially given the level of sophistication exhibited by disinformation campaigns around the world, most notably Russia’s interference in the 2016 American presidential election and its likely interference in the upcoming election. While Facebook and other social media giants have learned some lessons from the election interference of the last several years, the rapid pace of technological advancement ensures that the fight against disinformation will not end anytime soon.