Face-Swapping Technology: Dignity, Privacy & the First Amendment
New machine-learning technology is allowing even amateur video editors to conjure videos that convincingly replace people’s faces with those of others – frequently unwitting celebrities – to both creative and destructive ends. This digital face-swapping tech has been used for satirical internet videos and perhaps most famously to recreate a younger Princess Leia in the Star Wars film, Rogue One. In their most provocative form, these so-called “deepfakes” digital AI tools have been used to create X-rated content featuring the faces of popular Hollywood actresses grafted on to porn stars’ bodies. The videos have already engendered a firestorm that has led to bans on even freewheeling platforms like Reddit and Pornhub. This short presentation will explore whether the law can keep up with this controversial form of speech, and whether a balance can be struck to protect the reputational and privacy interests of unwitting subjects while upholding First Amendment principles.
• Do existing laws governing defamation, privacy, right of publicity, copyright, or the intentional infliction of emotional distress, or anti-revenge porn laws, protect the unwitting subjects of “deepfakes” videos?
• How does the legal analysis change when fake videos are passed off as real? When celebrities are involved?
• Will this technology make it harder to verify audiovisual content, and easier to generate fake news?
Jim Rosenfeld, Partner, Davis Wright Tremaine LLP