The Epistemic Threat of Deepfakes

Abstract

Deepfakes are realistic videos created using new machine learning techniques rather than traditional photographic means. They tend to depict people saying and doing things that they did not actually say or do. In the news media and the blogosphere, the worry has been raised that, as a result of deepfakes, we are heading toward an "infopocalypse" where we cannot tell what is real from what is not. Several philosophers (e.g., Deborah Johnson, Luciano Floridi, Regina Rini) have now issued similar warnings. In this paper, I offer an analysis of why deepfakes are such a serious threat to knowledge. Utilizing the account of information carrying recently developed by Brian Skyrms (2010), I argue that deepfakes reduce the amount of information that videos carry to viewers. I conclude by drawing some implications of this analysis for addressing the epistemic threat of deepfakes.

The Social Impact of Deepfakes

“…Can we believe any media that we see? The philosopher Don Fallis 18 refers to this as the epistemic threat of deepfakes. His argument flows from the power of visual media to carry information, which refers to how much signal is conveyed by a message.…”

The Distinct Wrong of Deepfakes

“…The available scholarship demonstrates a marked concern for the ethical implications of deepfake technology and its (potential) uses, as illustrated by a rising number of publications in law, information technology, communication studies, and political science Diakopoulos & Johnson, 2020;Fallis, 2020;Fletcher, 2018;Franks & Waldman, 2019;Maras & Alexandrou, 2019;Meskys et al, 2020;Rini, 2019;Silbey & Hartzog, 2019;Spivak, 2019;Vaccari & Chadwick, 2020;Westerlund, 2019). While these works importantly identify and signal the potentially detrimental consequences, notably less attention has been paid to the moral dimensions of deepfake technology and deepfakes themselves.…”

“…3 Such a warning would also offer a partial answer to concerns that deepfake technology is morally problematic because it undermines trust in the reliability of recordings. This risk is discussed by scholars who highlight that the harm of deepfakes should not be seen to lie solely in their ability to portray people as saying or doing things they did not necessarily say or do, but also in casting doubt on the reliability of visual and auditory sources as evidence of the actual occurrence of events Fallis, 2020;Fletcher, 2018;Maras & Alexandrou, 2019;Rini, 2019;Vaccari & Chadwick, 2020). This issue is brought up by Citron and Chesney (2019: 1785) who discuss the "liar's dividend" that follows from the wide circulation of deepfakes and makes it easier for liars to deny that they have done or said things when recordings show up.…”

“…Concerns have been raised about the potential use of deepfakes for blackmail, intimidation, and sabotage , ideological manipulation (Fletcher, 2018: 467), and incitement to violence . Broader implications in terms of trust, accountability, and plausible deniability have also been discussed Fallis, 2020;Fletcher, 2018;Maras & Alexandrou, 2019;Rini, 2019;Vaccari & Chadwick, 2020). While this literature performs an important task in identifying and signalling the potentially detrimental consequences of deepfake technology and assessing various technological, judicial, and administrative strategies for its regulation, notably less attention has been paid to the moral dimensions of deepfake technology and deepfakes themselves.…”

Huanlian, or changing faces: Deepfakes on Chinese digital media platforms

“…Researchers across disciplines have extensively discussed the philosophical, cultural, social, legal, and geopolitical implications of deepfakes. The scale at which this new kind of automated media manipulation can be deployed, along with the sophistication of the synthesized content, has led scholars to identify the emergence of "post-fact performance" (Fletcher, 2018) challenging established notions of authenticity (Floridi, 2018), eroding trust in the news (Vaccari and Chadwick, 2020), undermining the testimonial value of recorded media (Rini, 2020) and ultimately becoming an "epistemic threat" to societies (Fallis, 2020). A report by visual threat intelligence company Sensity correlates the rapid growth of deepfakes online-largely driven by pornography-with an increase in computer science research on the subject and the commodification of audiovisual synthesis tools (Ajder et al, 2019).…”