Taylor Swift has filed new trademark applications for two voice clips and one image that may be specifically designed to provide additional protection from artificial intelligence impersonation.
Swift’s applications were filed with the United States Patent and Trademark Office (PTO). Usually, trademarks cover names and logos. The sound marks filed by Swift cover a lesser-known category of trademarks called “sensory marks;” in some cases, smells and sounds can be trademarked, meaning that they can be protected as designating a particular source of goods or services. A sound mark would not give Swift ownership over her voice in the abstract. It would protect only a specific sound recording or a specific phrase or set of words spoken in a particular manner, with particular cadence, tone, and delivery. In other words, trademark protection for a sound mark is tied to the distinct phraseology and the way it is performed, not to the speaker’s voice as a whole.
One sound mark would protect Swift saying the phrase “Hey, it’s Taylor Swift, and you can listen to my new album, The Life of a Showgirl, on demand on Amazon Music Unlimited.” The second sound mark would protect the sentence “Hey, it’s Taylor. My brand-new album, The Life of a Showgirl, is out on Oct. 3, and you can click to pre-save it so you can listen to it on Spotify.”
The filings come as traditional copyright laws fail to guard against AI-generated content. AI can now generate completely new content that mimics an artist’s voice without outright copying.
The image that Swift is seeking to register is a photograph of her on stage at the recent Eras Tour in which she is holding a pink guitar in a sparkly leotard and knee-high boots.
Matthew McConaughey recently filed similar applications to protect his voice saying the iconic phrase “Alright, alright, alright” from his breakout film Dazed and Confused. In theory, if anyone used AI to generate a sound similar to McConaughey’s recording or Swift’s phrases, they could claim that the sounds violate their rights.