Taylor Swift Files to Trademark Her Voice and Image — The Legal Strategy Behind Her Fight Against AI Deepfakes
Taylor Swift Files to Trademark Her Voice and Image — The Legal Strategy Behind Her Fight Against AI Deepfakes
Taylor Swift has taken an unusual legal step to protect her identity in the age of artificial intelligence, filing trademark applications covering two audio clips of her voice and an image of herself performing. The move, reported in April 2026, reflects growing concern across the entertainment industry about the use of AI tools to create realistic audio and video featuring famous people without their permission.
The applications were filed with the United States Patent and Trademark Office through Swift’s company, TAS Rights Management. Two filings concern short audio phrases in which the singer introduces herself. One begins with the words “Hey, it’s Taylor Swift,” while the other uses “Hey, it’s Taylor.” A third application covers a specific photograph showing Swift holding a pink guitar while wearing a multicolored stage outfit.
The filings attracted attention because sound trademarks are less common than traditional trademarks involving names, logos and visual designs. They are used to identify the commercial source of a product or service through a recognizable sound. Applying this type of protection to a celebrity’s spoken voice raises interesting questions about how intellectual property law may adapt to new technologies.
The applications were interpreted by intellectual property attorney Josh Gerben as an effort to strengthen Swift’s protection against AI-generated impersonations. However, the legal filings themselves do not establish that every possible AI imitation would automatically violate trademark law. The exact scope of protection would depend on the applications, the decisions of the trademark office and the circumstances of any future dispute.
The concern is easy to understand. Modern AI systems can produce audio that closely resembles a real person’s voice, even when the person never recorded the words being spoken. Similar tools can generate images and videos that appear to show celebrities participating in events or making statements that never happened.
For public figures, the risks can be significant. A fabricated endorsement could make it appear that a celebrity supports a product they have never promoted. A manipulated political message could falsely suggest that an artist has taken a position on an issue. More disturbing examples involve fake explicit images or misleading videos that can spread rapidly before the person depicted has an opportunity to respond.
Swift has already been affected by the wider problem of unauthorized AI-generated content. In early 2024, sexually explicit fake images depicting her circulated online, generating widespread criticism and renewed calls for stronger protections against non-consensual deepfakes.
The trademark applications therefore represent part of a broader effort to address how a person’s identity can be used commercially and digitally. Traditional copyright law may protect particular recordings or photographs, but AI-generated imitations can create new legal complications when they reproduce a person’s voice or appearance without directly copying an existing recording.
Trademark protection may provide another possible legal route, although its effectiveness in this area remains uncertain. A sound mark generally needs to function as a recognizable identifier of a particular source. Whether short spoken phrases can provide meaningful protection against sophisticated AI impersonations is a question that legal experts will continue to examine.
Swift is not the only public figure exploring this approach. Actor Matthew McConaughey has also pursued trademark protections involving his voice and recognizable expressions. The growing interest among celebrities suggests that traditional intellectual property tools may be adapted as the entertainment industry responds to AI-generated content.
The issue extends beyond famous performers. Musicians, actors, journalists and ordinary individuals can all become targets of fabricated audio and images. The technology makes it easier to create convincing material, while social media platforms allow it to reach large audiences in a short period of time.
For the music industry, the question is especially important because a singer’s voice is one of the most recognizable elements of their professional identity. Fans associate particular vocal characteristics with specific artists, and companies may attempt to use AI-generated versions of those voices to produce music, advertisements or other commercial content.
Swift’s filings highlight the importance of consent. An artist may choose to experiment with technology, authorize a particular collaboration or allow their voice to be used in a specific project. That is different from having someone create an imitation without permission and distribute it as though it were genuine.
The legal process is not a complete solution to the deepfake problem. Even if a trademark application is approved, enforcement can be difficult when content is produced anonymously or distributed across multiple platforms. Copyright, publicity rights, privacy laws and platform policies may all play a role in future disputes.
Nevertheless, Swift’s decision has drawn attention to a developing area of intellectual property law. By seeking protection for both spoken phrases and a visual representation of herself, she is exploring ways to defend elements of her identity that have become increasingly vulnerable to digital imitation.
The applications may also influence how other performers approach their own legal protections. As AI tools become more accessible, the entertainment industry will likely face more disputes over the unauthorized use of voices, faces and recognizable personal characteristics.
Swift’s move is therefore more than a legal filing involving a famous singer. It is part of a much wider debate about who controls a person’s digital identity and what protections should exist when technology makes it possible to imitate someone with increasing accuracy.
Source: Reuters, Associated Press, The Guardian, CBS News