Regulating deep fakes: legal and ethical considerations

作者：Edvinas Meškys, Aidas Liaudanskas, Julija Kalpokienė, Paulius Jurčys · 发表于：Journal of Intellectual Property Law & Practice · 年份：2019 · DOI：10.1093/jiplp/jpz167 · 被引用次数：187 · 研究领域：FinTech, Crowdfunding, Digital Finance

On 29 September 2019, ZaoApp was introduced in China via iOS Store. ZaoApp is a face-swapping app that uses clips from a great variety of films and TV shows, convincingly changing a character’s face by using selfies from the user’s phone.1 The results were so exciting that, within three days, ZaoApp was the most downloaded app in China. It has been the most successful adoption of the ‘deep-fake’ technology so far, allowing a video to be generated in under eight seconds from one single photo, video clip or GIF file. Soon after the release of ZaoApp, online payments operator Alipay, which has approximately 1 billion active users, released a warning that its users should ‘[r]est assured that no matter how sophisticated the current facial swapping technology is, it cannot deceive our payment apps’.2 The technology behind deep fakes is rapidly improving. The democratization of deep fakes through publicly available apps raises a number of legal, social and ethical questions. Is a legal intervention necessary? Should deep fakes be banned completely? If regulators were to step in, what would be the most efficient way to steer how online communities create deep fakes? Should any obligations be imposed on platform operators? What is the role of intellectual property in addressing this novel phenomenon? If deep fakes detection technology is created, who should be entitled to control it: governments, platforms, or others altogether? What is the best regulatory approach to deep fakes in l...