The Authors' Take - Fake it till you make it: an examination of the US and English approaches to persona protection as applied to deepfakes on social media

Fake it till you make it: an examination of the US and English approaches to persona protection as applied to deepfakes on social media

by Frederick Mostert and Emma Perot

In both the UK and US, deepfakes are likely to be addressed by current law. In the UK, passing off, while more complex than the right of publicity, could be helpful to address deepfakes where they are used in a way that conveys endorsement. For non-endorsement uses, defamation might be grounds for legal action. In the US, the right of publicity could be deployed. However, these causes of action inadequately address the potential of deepfakes to sway public opinion, damage the reputation of the individual depicted, and impinge on privacy and dignity of the individual if used in pornography. It may be challenging to track down the uploader, and the delay between commencing legal action and receiving an injunction can be too late if the deepfake goes viral. The potential velocity and amplification of deepfakes on the web allows for nefarious uses of this relatively new technology.

While the creation of rules in this regard are useful, there is also a need for technology to facilitate content moderation. As deepfake videos become increasingly difficult to distinguish from real footage, news organisations such as Reuters are conducting their own deepfakes experiments to stay one step ahead of the fake news disseminators. Detection of deep fakes is still largely an open research problem. Deepfakes are already approaching the limits of human perception to discern images. Automated techniques have focused on heuristics such as blinking patterns or incongruities in perspective and shading. It is unlikely that these will remain effective for very long. The arms race will continue for the foreseeable future, and as such detecting deepfakes may well remain an aspirational goal rather than something that can be relied upon as a solution.

[This is an Authors' Take post, which provides readers with an insight into current IP scholarship, featuring preliminary comments and thoughts from authors of articles accepted for publication in forthcoming issues of the Journal of Intellectual Property Law & Practice (OUP). The full text of this contribution, which will be included in our special Image Rights Issue, will be made available on Advance Access soon]

The Authors' Take - Do Deepfakes Pose a Golden Opportunity? Considering Whether English Law Should Adopt California's Publicity Right in the Age of the Deepfake

Do Deepfakes Pose a Golden Opportunity? Considering Whether English Law Should Adopt California's Publicity Right in the Age of the Deepfake

by Kelsey Farish

In 2017, a machine learning algorithm was shared on Reddit as a tool to insert faces of celebrity actresses into pornographic videos. This “deepfake” phenomenon has since spread across social media, and is no longer confined to sexual contexts. The technology can be used to swap faces in film scenes, or even digitally insert people into their favourite movie clips. Although the results are often comical, deepfake sophistication and realism has rapidly improved over the last two years, making them difficult to spot as fake. There is a growing concern that such videos could be used to extort, intimidate, or otherwise defame an individual. In such instances, could the victim portrayed in the deepfake bring a lawsuit against its creator?

In California, perhaps. There, a person has a statutory and common law “publicity right”, which is a cause of action used to prevent or penalise any misappropriation of one’s image, photograph, or voice. By contrast, the lack of a recognised image right under English law can be a source of frustration amongst claimants, and debate amongst lawyers. Drawing upon her knowledge of English and Californian law, the author explores whether or not California’s codified publicity right is superior to that of the English piecemeal approach, using the deepfake phenomenon as a case study.

It is hoped that the article will be of interest to practitioners and academics involved in technology and intellectual property rights matters. Those working in artificial intelligence and digital media may also find the article insightful, as it aims to explain the legal landscape surrounding this disruptive new technology. Finally, given the role our selfies and snaps play in the internet ecosystem, this discussion on image rights in the age of the deepfake may be interesting to the casual reader as well.

[This is an Authors' Take post, which provides readers with an insight into current IP scholarship, featuring preliminary comments and thoughts from authors of articles accepted for publication in forthcoming issues of the Journal of Intellectual Property Law & Practice (OUP). The full text of this contribution will be made available on Advance Access soon and included in the 2020 JIPLP Special Image Rights Issue]