Taylor Swift’s Management Seeks Trademarks for Voice Clips and Onstage Image to Combat AI Deepfakes

Taylor Swift’s management company, TAS Rights Management, recently submitted three trademark applications to the U.S. Patent and Trademark Office (USPTO), according to reports from CBS News and Reuters. Two of these applications seek protection for short audio clips featuring Swift’s voice uttering phrases such as “Hey, it’s Taylor Swift” and “Hey, it’s Taylor.” The third application covers an image of Swift performing onstage with a guitar, as detailed by CBS and Reuters.

NBC News published an example from the filings that includes a longer promotional audio sample referencing one of Swift’s albums and a streaming platform. These filings represent a proactive legal measure aimed at limiting the unauthorized use of Swift’s voice and likeness, particularly in the context of AI-generated deepfakes that could mislead consumers or imply false endorsements.

Intellectual property attorney Josh Gerben, quoted by CBS News, characterized Swift as a “leader in the intellectual property space” and predicted that other public figures may follow suit with similar trademark filings. This trend reflects growing celebrity efforts to leverage trademark law as a defensive mechanism against AI-driven impersonations.

From a legal perspective, trademark law traditionally protects source-identifying signals and prevents consumer confusion in commercial contexts. Meanwhile, state publicity-rights statutes address unauthorized commercial exploitation of a person’s name or likeness. Copyright law protects fixed creative works such as songs or recordings but does not neatly cover an actor or singer’s voice or synthetic likenesses that are not direct copies of copyrighted material. These distinctions are significant because each legal framework involves different burdens of proof and remedies.

Public reporting frames Swift’s trademark filings as part of a broader celebrity response to the rise of AI deepfakes. CBS News noted that actor Matthew McConaughey has filed related trademarks in the same period, underscoring a wider industry movement. However, trademark filings are viewed as a partial solution since they require a commercial-use nexus and a likelihood of consumer confusion to be enforceable.

For AI developers, product managers, and legal teams, these filings highlight friction points between synthetic media technologies and existing intellectual property frameworks. Voice-cloning and generative-image system creators should be aware that public figures are actively pursuing trademark and publicity rights remedies, which may influence takedown demands, content moderation policies, and platform liability assessments.

Legal teams advising platforms will closely monitor how the USPTO processes these applications and any subsequent office actions. Additionally, the evolution of case law will be critical in determining whether registrations for voice clips or likenesses create enforceable limits on generative model outputs.

Industry observers should track three key indicators: the USPTO’s response to these trademark applications; enforcement actions initiated by trademark holders or rights managers related to AI-generated content; and developments in state publicity-rights cases or federal litigation addressing synthetic voice and likeness claims. Platform policy updates, automated synthetic media detection tools, and notice-and-takedown procedures will be immediate operational levers to watch.

Overall, reporting from CBS News, NBC News, Reuters, and commentary in The Conversation frames these filings as a pragmatic and incremental legal response to the challenges posed by deepfakes. While these trademark applications signal that rights holders are testing available intellectual property mechanisms in the AI era, open questions remain regarding statutory fit, cross-jurisdictional enforcement, and the practical scope of trademark claims against algorithmically generated content.