Taylor Swift Seeks to Trademark Her Image: Here’s Why TikTok Deepfake Ads Highlight the Importance.

Last week, Taylor Swift submitted three trademark applications to safeguard her image and voice. One application focuses on a famous photo of the pop star holding a pink guitar during her record-setting Eras tour, while the other two are sound trademarks for the phrases: “Hey, it’s Taylor Swift” and “Hey, it’s Taylor.”

This action comes as AI-generated deepfakes become increasingly prevalent on social media. Individuals risk having their likenesses misused in nonconsensual AI-created content; earlier this month, an Ohio man became the first to be convicted under a new federal law against “intimate” visual deceptions. Celebrities, too, face threats from both explicit deepfakes and misleading endorsements.

A recent report from AI detection firm Copyleaks reveals that Swift and other celebrities have had their images used in unscrupulous advertisements. Researchers found a series of sponsored videos on TikTok that seemed to feature Swift, Kim Kardashian, Rihanna, and others endorsing “potentially fraudulent or harmful services,” employing what the researchers describe as “realistic-sounding voices” and “textured filters designed to conceal imperfections in the AI-generated visuals.”

The fake advertisements depict Swift and others in typical interview settings—like red carpet events or talk shows. Instead of responding to questions, however, these AI-generated versions of the celebrities promote alleged rewards programs in which TikTok users can earn money by providing feedback on content presented to them.

“I was reading about digital behavior this week and came across a testing feature called TikTok Pay,” a deepfaked version of Swift states in an ad that manipulates footage from her real appearance on The Tonight Show Starring Jimmy Fallon last October. “Certain users are being invited to watch videos and submit opinions.” The deepfaked Swift adds that the program is currently in a “limited rollout” and encourages viewers to check if they qualify, saying: “If the page opens for you, don’t overthink it.”

Unsurprisingly, anyone who clicks is approved. These ads ultimately direct the user to a third-party service that, despite featuring the TikTok branding, is evidently designed using the AI platform Lovable, whose branding appears both on the page and in the URL. At this point, researchers note that users are prompted to enter their name and personal details.

While it remains unclear what the advertisers intend to do with the data collected through their celebrity deepfake campaigns, scam advertisements with similar goals are alarmingly common. Last week, the nonprofit Consumer Federation of America filed a lawsuit against Meta, claiming that the tech company misled Facebook and Instagram users regarding its efforts to combat scam ads while profiting from their spread. On Monday, the US Federal Trade Commission reported a significant rise in social media scams, with Facebook scams leading in total financial losses.

It’s no wonder that Swift and her colleagues are pursuing legal measures to dissociate themselves from this fraudulent landscape. Although Swift hasn’t publicly addressed her trademark filings’ rationale, the potential reputational harm that deceptive deepfakes could inflict on her billion-dollar brand is impossible to ignore. The challenge is that these fraudulent practices are becoming increasingly advanced with each passing day.