Should Taylor Swift’s Fame Get Faster Protection?

As generative artificial intelligence (AI) becomes more commercially powerful, trademark law is quietly emerging as one of the most effective tools for controlling digital identity online. Celebrities and major corporations increasingly leverage trademark portfolios, platform relationships, and licensing infrastructure to shut down AI misuse and monetize synthetic versions of their identities. Ordinary creators who may be building their audience are left relying on fragmented copyright claims and inconsistent state-law remedies without the resources to protect their most precious asset- their name and likeness. The result is a growing two-tier system: those with fame, money, and leverage are best positioned to control how AI replicates them while everyone else is left behind. If AI can replicate anyone, but only the famous can realistically stop it, the law is no longer protecting identity equally.

Matthew McConaughey and Taylor Swift are prime examples of individuals who are afforded the greatest protection and are also best positioned to monetize their likeness. McConaughey secured multiple trademarks covering his catchphrases, most notably his “Alright, alright, alright!” from Dazed and Confused, to create more legal leverage against AI misuse. Reportedly in response to concerns about AI fakes, Swift filed trademark applications covering short voice phrases such as “Hey, it’s Taylor Swift,” and “Hey, it’s Taylor.” Swift also filed an application for a photograph of herself holding a pink guitar with a black strap and wearing a multi-colored bodysuit with silver boots. Even if approved, however, these efforts may not stop every deepfake because trademark rights remain limited to source-identifying uses in commerce. That said, they create a faster, more credible path to federal claims, platform takedowns, and licensing negotiations.

HOLLYWOOD, CALIFORNIA – MARCH 10: Matthew McConaughey attends the 96th Annual Academy Awards at Dolby Theatre on March 10, 2024 in Hollywood, California. (Photo by Jeff Kravitz/FilmMagic)

FilmMagic

This two-tier reality is not merely an inconvenience; it is a structural problem embedded in the law itself. Trademark law rewards those who can document ownership, prove consumer recognition, and threaten federal enforcement. While famous brands and celebrities clear these hurdles with ease, ordinary creators often cannot prove harm until after the damage is done. The same disparity shapes how platforms respond to AI misuse.

YouTube’s Dream Track experiment, for example, allowed selected users to generate AI music clips using the voices of artists, such as Demi Lovato, John Legend, and T-Pain, with artist consent and platform approval. YouTube has also built likeness-detection tools, first in partnership with CAA and later for broader groups of creators and entertainment figures, to help eligible individuals find AI deepfakes and request removal. But the problem begins with eligibility: the first, best, and most automated protections tend to arrive for celebrities, major creators, and rights holders with market leverage.

The pattern extends beyond platform tools to legal enforcement itself. In copyright, large rightsholders possess the infrastructure to pressure platforms quickly and effectively. This same dynamic also extends to claims of copyright infringement. For example, Disney sent Google a cease-and-desist letter alleging that Google’s AI services were generating and distributing Disney characters, including characters from Frozen, Moana, Toy Story, Marvel, Star Wars and The Simpsons. Disney also demanded that Google remove flagged YouTube videos and implement safeguards to prevent future AI outputs using Disney-owned characters from being uploaded. By the next day, links to some videos redirected to a message saying the video was unavailable because of a copyright claim by Disney.

These examples show the broader structure of power that copyright and trademark law reinforces. Major rightsholders with famous characters, registered brands, and established enforcement channels can get rapid results without first winning a lawsuit. In contrast, an average illustrator whose style is copied by an AI video or image generator cannot point to a globally recognized brand portfolio or a stable of registered marks. These individuals rely on copyright complaints, platform reporting tools, or state-law “right of publicity” claims, which vary by state and often require more time, money, and proof. In fact, a single user who believes they have been a victim of infringement is frequently labeled delusional or crazy. They must often seek legal representation that is willing to take on their matter on contingency (where payment can only be in success of a settlement or major positive outcome to litigate). The disparity between the perception that a celebrity’s right to claim theft has more value than a lone creator is the real example of ‘intellectual property gentrification’ among creator ‘have-nots.’

Legislative efforts are beginning to address this imbalance. The proposed NO FAKES Act would create federal protections for a person’s voice and visual likeness against unauthorized digital replicas, complete with notice-and-takedown procedures and preemption of conflicting state laws. Tennessee’s ELVIS Act already points in this direction, extending publicity-right protections to voice and targeting unauthorized AI imitation. These laws are imperfect because they must still safeguard parody, criticism, and other protected speech, but they recognize the core issue: identity protection should not depend on whether you are Disney, Taylor Swift, or someone without a legal department.

A more effective legislative solution would combine the federal rights-based framework of the NO FAKES Act with affirmative structural mechanisms designed to lower enforcement barriers for ordinary creators. Specifically, a potential solution would mandate that platforms hosting AI-generated content implement universal identity-verification tools, not merely for celebrities, but for any individual who registers a verified digital likeness with a designated federal registry. Such a registry could function analogously to the Copyright Office’s or United States Patent and Trademark Office’s registration system, providing a formal record that triggers statutory damages upon infringement, thereby eliminating the need for creators to independently prove consumer confusion or commercial harm before obtaining relief.

To be clear, celebrities and corporations should not lose the ability to protect themselves from AI misuse. Rather, the problem is that the current system concentrates the strongest tools in the hands of those who already enjoy the greatest economic and legal advantages. As AI-generated content proliferates, trademark law risks becoming a gatekeeping mechanism that determines whose identity can realistically be protected, licensed, and enforced online. Without broader, more accessible protections for independent creators, the gap between institutional rightsholders and everyone else will only widen. The question is not whether this gap exists, but how long it will take before the average creator’s work, voice, or likeness is copied, monetized, and forgotten.

By Elsa Ramo and Michael Segal