Law
Understanding Deepfake Defamation and Reputational Torts
Quick fact
A 2019 report by Deeptrace Labs found that deepfake technology had been used to create 15,000 fraudulent videos, mostly non-consensual pornography, which can also harm reputations. Courts are only beginning to develop clear legal standards for such harm under existing defamation law.
Why this is interesting
Imagine a video of a public figure saying something they never said, so realistic that millions believe it. If it damages their reputation, what legal recourse do they have?
Read the full explanation
Understanding Understanding Deepfake Defamation and Reputational Torts
Deepfakes are synthetic media produced by artificial intelligence, making it appear that someone said or did something they never did. Under traditional defamation, a plaintiff must show that a false statement was published to a third party, that it was defamatory, and that it caused harm to reputation. Deepfakes fit this framework because the AI-generated video is a 'statement'—a representation of fact—that is published when shared or streamed. What makes deepfakes particularly insidious is their realism, which makes it hard for audiences to identify them as false, and for victims to prove falsity and source.
A deeper explanation
The legal mechanism hinges on whether a deepfake qualifies as a 'defamatory statement.' If the deepfake portrays the person in a false and damaging light, it can be defamatory if a reasonable person would understand it as depicting facts about the person. However, the Constitution imposes limits, especially for public figures, who must show 'actual malice'—knowledge of falsity or reckless disregard for the truth—per New York Times v. Sullivan. Deepfakes complicate this because the creator may be anonymous, and the publisher (e.g., a social media platform) may be protected under Section 230 of the Communications Decency Act, which immunizes platforms from liability for user-generated content. Additionally, deepfakes might give rise to other reputational torts such as 'false light'—publicity that places a person in a false light highly offensive to a reasonable person—and intentional infliction of emotional distress, if the deepfake causes severe emotional harm. The challenge for courts is to adapt these doctrines to AI-generated content, determining the degree of fault, proving damages, and identifying the responsible party.