In recent times, the menacing evolution of deepfake technology has paved the way for a dystopian digital landscape, where discerning reality from fabrication becomes a daunting task. Deepfake technology, powered by sophisticated AI and machine learning algorithms, enables the creation of hyper-realistic videos or images by grafting an individual’s face or head onto someone else’s body. The implications of this technological marvel are as boundless as they are harrowing, especially when misused for creating explicit material without the consent of the individuals depicted.
The victims of malicious deepfakes are often high-profile female celebrities and internet personalities. Their immense online following and recognition makes them prime targets for exploitational deepfakes that aim to go viral. TikTok stars like Addison Rae Easterling, Charli D’Amelio, and Bella Poarch have been subjected to deepfakes featuring their likeness in sexually explicit videos without consent. In April 2022, a deepfake depicting Easterling in a pornographic video amassed over 19 million views on an adult site, bringing the issue to mainstream attention. D’Amelio and Poarch have also been targeted, with Poarch tearfully addressing a deepfake sex video that circulated online. These poignant examples showcase the disconcerting scale and virality of malicious deepfakes targeting prominent female internet celebrities.
The emotional trauma and reputational damage that such deepfakes elicit is immense. Victims describe feelings of powerlessness and rage at having their likeness appropriated and misused in dehumanizing ways. The intimate violation of privacy through deepfakes can also lead to mental health issues like anxiety, depression, and post-traumatic stress disorder. Easterling described the toll of the explicit deepfake as “unbearable”, leading her to question her self-worth. The jarring sense of exploitation leaves deep psychological scars.
Deepfakes featuring celebrities also raise legal concerns around copyright and right to publicity. Some victims pursue legal routes to bring down the content. Poarch’s team filed lawsuits over the deepfake video misusing her likeness. Easterling sent cease-and-desist letters to sites hosting the deepfake. However, the decentralized nature of the internet poses challenges for enforcing legal measures against malicious deepfakes that spread like wildfire online. Victims are often forced to play an endless game of whack-a-mole.
The diffusion of explicit deepfake material occurs rapidly through social media channels and lesser-regulated corners of the internet. Despite most major platforms like Twitter, Reddit, and Pornhub instituting policies banning nonconsensual synthetically generated media, the sheer volume across user-generated content makes it impossible to catch all policy violations. For instance, Twitter removed the Easterling deepfake only after it had become massively viral, showcasing the struggle of platforms to effectively monitor a growing issue. The inconsistent enforcement gives policy loopholes for deepfakes to slip through and propagate online.
The dark web provides an even more lawless avenue for malicious deepfakes, beyond the purview of platform guidelines. Sites on the dark web explicitly allow nonconsensual deepfakes and have virtually no restrictions against revenge porn or exploitation. For instance, asubdomain on the dark web called “deepfakesweb” emerged in 2018, exclusively hosting thousands of AI-generated fake porn videos of celebrities and civilians, predominantly women. Such unregulated corners of the internet allow deepfakes depicting sexual abuse to run amok.
Advocacy groups have sprung up to provide support for deepfake victims, like the Coalition Against Stalkerware. They provide counseling and legal resources to help victims fight back against the exploitation. Anti-deepfake software tools are also emerging to detect AI-manipulated media, like Microsoft’s Video Authenticator. While still a work in progress, such technology aims to curb deepfakes at the source. There are also growing calls for comprehensive regulations against revenge porn and nonconsensual deepfakes.
In September 2022, California passed the global first anti-deepfake law imposing hefty fines for creating and distributing malicious deepfakes without consent. The law signals a move towards holding bad actors accountable and protects residents from exploitative deepfakes. However, critics argue the law may be difficult to enforce given the viral spread of deepfakes online across borders. Meaningful progress requires coordinated action between governments, tech companies, researchers, and advocacy groups.
As we venture deeper into the digital age, the battle against deepfake technology and its malicious uses epitomizes the broader struggle to maintain truth, privacy and consent in the virtual realm. The harrowing tale of deepfake victims serves as a stark reminder that the digital frontier, while ripe with innovation, also harbors perils that demand our immediate attention and action. If left unchecked, deepfakes threaten to poison public discourse and undermine the social fabric. The choices we make today will determine whether technology elevates or degrades our shared humanity.