For decades, the battle over a celebrity’s image was fought in the realm of “right of publicity”—a patchwork of state laws designed to stop a company from putting a movie star’s face on a billboard without permission. But the arrival of generative AI has fundamentally shifted the stakes. We are no longer just dealing with unauthorized photos, but with “digital twins”: AI-generated voices that can sing any song and deepfake videos that can say anything, often indistinguishable from the real person.
In response, a new legal vanguard is emerging. High-profile stars, most notably Taylor Swift, are pivoting toward trademark law to protect their identities. While copyrights protect a specific song or a specific photo, trademarks protect a “brand.” By attempting to trademark their voices and likenesses, celebrities are treating their very existence as a commercial product, creating a federal shield that is often faster and more potent than waiting for a civil lawsuit to wind through the courts.
This shift isn’t just about vanity or vanity-driven litigation. it is a strategic move to seize control of the “training data” that fuels AI. When a voice is trademarked, the unauthorized use of that voice in an AI model can be framed not just as a privacy violation, but as trademark infringement—a charge that allows for more aggressive takedown notices and statutory damages.
The Swift Strategy: Treating Identity as an Asset
Taylor Swift has long been a master of intellectual property, but her recent moves to secure trademarks for her voice and image represent a sophisticated evolution in artist rights. As reported by Quarles and Pinsent Masons, these filings are designed to combat the proliferation of AI clones that can mimic her signature cadence and tone.
The logic is simple: if “Taylor Swift’s voice” is a registered trademark, any AI company that offers a “Swift-style” voice generator is potentially infringing on a federally protected brand. This gives her legal team a powerful tool to issue cease-and-desist letters to platforms hosting deepfakes, effectively treating a fake audio clip as a counterfeit product rather than a mere piece of misinformation.
This approach is becoming a global blueprint. From the U.S. To the high-glamour courts of Bollywood, stars are increasingly turning to civil litigation to fight deepfakes. According to The Conversation, the trend reflects a growing desperation among public figures who find that traditional defamation laws are too sluggish to stop a viral AI clip from damaging their reputation in real-time.
Can the Average Person Trademark Their Face?
The central question for the rest of us is whether this “celebrity shield” is accessible to the general public. As The Washington Post explores, the answer for most people is a resounding “probably not.”
Trademark law is not designed to protect privacy; it is designed to protect consumers from confusion in the marketplace. To successfully register a trademark with the U.S. Patent and Trademark Office (USPTO), an applicant must demonstrate “use in commerce.” This means the mark must be used to identify goods or services. For Taylor Swift, her voice is inextricably linked to her commercial output—albums, tours, and endorsements. For a software engineer or a teacher, their voice and image are not “brands” used to sell products in a commercial market.
the cost and complexity of maintaining a trademark portfolio are prohibitive for the average citizen. While a celebrity can afford a team of IP lawyers to monitor the internet for infringements, a private individual would find the process an expensive exercise in futility, as they lack the commercial standing to satisfy the USPTO’s requirements.
Comparing Protections: Right of Publicity vs. Trademarks
To understand why celebrities are making this move, it is helpful to look at the different legal mechanisms available for protecting one’s identity. While the “Right of Publicity” has been the standard, trademarks offer a different set of advantages.
| Feature | Right of Publicity | Trademark Law |
|---|---|---|
| Legal Basis | State law / Common law | Federal law (USPTO) |
| Primary Goal | Privacy & Personal Control | Consumer Protection/Brand Integrity |
| Requirement | Recognizable identity | Use in commerce |
| AI Application | Sue for damages after use | Preemptive takedowns & infringement |
| Accessibility | Available to most citizens | Reserved for commercial brands |
The Legislative Gap and the Road Ahead
The rush toward trademarks is a symptom of a larger problem: the law is lagging behind the technology. Currently, there is no comprehensive federal law in the U.S. That explicitly prohibits the non-consensual creation of an AI clone. While some states have passed “deepfake” laws—often focused on non-consensual explicit imagery—the broader issue of “identity theft” via AI remains a gray area.
Industry advocates and lawmakers are currently debating the “NO FAKES Act,” a proposed federal bill that would create a property right in one’s voice and likeness, effectively giving everyone—not just celebrities—a federal tool to fight AI clones. If passed, this would eliminate the need for the “use in commerce” hurdle required by trademark law, providing a streamlined path for individuals to protect their digital selves.
Until such legislation is enacted, the “Swift Model” remains the gold standard for those with the means to implement it. For the rest of us, the best defense remains digital hygiene: being mindful of the biometric data we share online and utilizing platform-specific reporting tools to flag AI-generated impersonations.
Disclaimer: This article is for informational purposes only and does not constitute legal advice. For specific legal concerns regarding trademarks or identity rights, please consult a licensed intellectual property attorney.
The next major checkpoint for this legal evolution will be the continued deliberation of the NO FAKES Act in Congress, as lawmakers weigh the balance between individual identity rights and the First Amendment protections of AI developers. We will continue to track the progress of this legislation and any subsequent USPTO rulings on “persona” trademarks.
What do you think? Should your digital identity be a legal property right? Share your thoughts in the comments below.
