Taylor Swift Files New Trademarks as Artists Seek Protection Against AI Replication
Taylor Swift has taken a new legal step that reflects a growing concern across the entertainment industry: how to protect an artist’s identity in an era increasingly shaped by artificial intelligence.
According to recent filings with the United States Patent and Trademark Office, Swift’s company, TAS Rights Management, submitted three trademark applications on April 24. Two of those applications focus on sound marks tied directly to her voice, while a third centers on a specific visual representation of the artist.
The sound trademarks include the phrases “Hey, it’s Taylor Swift” and “Hey, it’s Taylor”, both of which aim to secure legal protection over recognizable audio identifiers associated with the singer. The third filing describes a detailed image of Swift holding a pink guitar with a black strap, dressed in a multi-colored iridescent bodysuit and silver boots, standing on a pink stage in front of a multi-colored microphone with purple lighting. By formalizing this imagery, the filing attempts to define a distinct visual signature that could be protected under trademark law.
The applications were first identified by intellectual-property attorney Josh Gerben, who has noted that such filings may signal a broader shift in how public figures defend their likenesses. While trademark law has traditionally been used to protect brand identifiers such as logos or slogans, some legal experts are now exploring whether it can also serve as a tool against unauthorized AI-generated content.
Swift’s move follows a similar strategy pursued by Matthew McConaughey, whose legal team successfully secured multiple trademarks in 2025. Among them was a sound mark tied to his well-known phrase “Alright, alright, alright”, originally delivered in the 1993 film Dazed and Confused. That approach has been described by some attorneys as an attempt to expand the legal options available when confronting digital impersonation.
At the state level, laws in places such as California and New York already offer protections through so-called “right of publicity” statutes, which prohibit the unauthorized commercial use of a person’s name, image, or likeness. However, trademark claims can be brought in federal court, potentially offering a broader and more uniform legal pathway. This distinction is seen as particularly relevant as AI-generated content often circulates across jurisdictions, complicating enforcement under state-specific laws.
Swift’s filings also come after several instances in which her likeness has been used without authorization in AI-generated material. These include manipulated images and audio, as well as content produced through platforms such as Meta’s AI tools. In one widely circulated case during the 2024 U.S. presidential election cycle, AI-generated images falsely suggested that Swift had endorsed Donald Trump, highlighting how such technology can be used to misrepresent public figures.
Legal analysts suggest that securing trademarks tied to voice and visual identity could provide an additional layer of protection. For example, if an AI system produces audio that closely mimics Swift’s voice using a phrase covered by a registered trademark, her legal team could argue trademark infringement alongside other claims. Similarly, protecting a defined visual image could help challenge AI-generated pictures that evoke her likeness in recognizable ways.
Despite the potential, the effectiveness of this strategy remains largely untested in court. Trademark law has not historically been applied to protect a person’s general identity, and its application to AI-related disputes is still evolving. However, recent actions by major companies suggest that trademark enforcement may play a role in shaping how AI platforms operate. In December 2025, Disney issued a cease-and-desist letter to Google over alleged misuse of its intellectual property through the Gemini AI system, prompting the removal of disputed content.
