惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

H
Help Net Security
月光博客
月光博客
IT之家
IT之家
B
Blog RSS Feed
T
Tailwind CSS Blog
The GitHub Blog
The GitHub Blog
博客园 - 三生石上(FineUI控件)
MyScale Blog
MyScale Blog
J
Java Code Geeks
Stack Overflow Blog
Stack Overflow Blog
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
博客园 - Franky
博客园 - 叶小钗
阮一峰的网络日志
阮一峰的网络日志
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
U
Unit 42
博客园_首页
B
Blog
V
V2EX
腾讯CDC
Vercel News
Vercel News
量子位
Microsoft Security Blog
Microsoft Security Blog

Fortune | FORTUNE

One man can kill Bill Ackman’s $64 billion bid for Universal Music Group—and no one knows what he’ll do | Fortune Poppi’s cofounder pitched her startup on Shark Tank while 9 months pregnant and landed a $400,000 deal—now it's worth $2 billion | Fortune Teen boys are choosing AI girlfriends over real ones for 'maximum control, zero rejection'—experts say it could make them unemployable | Fortune A United American merger is by no means impossible given the president 'loves big deals' | Fortune Reed Hastings’s planned exit from $455 billion Netflix ‘had nothing to do with’ the failed deal for Warner Bros., says Ted Sarandos | Fortune Meet Joe McCann: The high-flying crypto trader held in Tanzania after sudden death of his influencer fiancée Ashly Robinson | Fortune Gen Z is carving a different path in the housing market by doing it alone | Fortune U.S. Catholic leaders criticize Trump for ‘disparaging words’ about the pope as Vatican clash risks alienating Catholic voters | Fortune China has ‘nearly erased’ America’s lead in AI—and the flow of tech experts moving to the U.S. is slowing to a trickle, Stanford report says | Fortune Self-made millionaire behind $5 billion Skims Emma Grede says it all began with a cold call to Kris Jenner: Emma Grede—the self-made millionaire behind the $5 billion Skims empire—says it all began with an audacious cold call to Kris Jenner: ‘The difference between me and someone else is, I made it happen’ | Fortune Americans have never been this gloomy about the economy. Wall Street has never cashed in harder | Fortune ‘The college grading system [is] almost meaningless’: People see the Ivy League as an easy A and with flawed admissions standards | Fortune The CEO of $8.5 billion Japanese car giant Nissan plays the drums in a band and hits the tennis courts to destress from the top job | Fortune New York governor's take on a millionaires tax: fancy pied-à-terre second apartments worth over $5 million | Fortune Pope Leo XIV: A ‘handful of tyrants’ are ravaging earth with war and exploitation | Fortune Trump has no plan to cut the $39 trillion national debt, but he does want to cut childcare. His budget director is scrambling to clarify | Fortune China's economy grows 5% in first quarter, surprising economists to the upside | Fortune Everyone was wondering what Trump wanted more: Warsh smoothly seated at the Fed, or for Powell to pay. We have our answer | Fortune Palantir exec: the biggest mistake retailers are making with AI? Trying to do it all with one agent | Fortune American YouTuber who calls himself a 'troll' sentenced to 6 months in Korean prison for literally dancing on wartime graves | Fortune BBC plans to cut up to 2,000 jobs to save 10% of annual budget | Fortune Canva debuts a new suite of agentic tools, as the design app quietly becomes one of the world’s most used AI services | Fortune Moody's CEO: AI has a trust problem – better models won’t fix it | Fortune Top New York surgeon: Americans have better data for choosing restaurants than surgeons. That has to change | Fortune The Iran war’s fertilizer shock is hammering American farmers, and 70% can’t afford what they need for this year’s growing season | Fortune Education experts to Mamdani: Why are you foisting AI on our kids? | Fortune This CEO pirated video games as a teen and became a hacker for the Air Force. Now he’s built a $3 billion cyber firm | Fortune Teacher, blame thyself: Yale report savages Ivy League schools for destroying American trust in higher education | Fortune Fed chair nominee Kevin Warsh is worth more than $100 million and has stakes in SpaceX and Polymarket | Fortune From wool sneakers to GPUs: Allbirds’ desperate AI pivot and 600% stock surge, explained | Fortune
Taylor Swift just exposed a blind spot in AI law — and it...
The Conversation · 2026-05-30 · via Fortune | FORTUNE

As one of the most popular celebrities in the world, Taylor Swift has already endured her share of AI-related abuse.

Fake nudes of the singer have spread widely online. Her voice and likeness have also been used to create fabricated political messages and bogus product endorsements.

In April 2026, Swift pushed back. Her intellectual property and brand management company, TAS Rights Management, filed trademark applications covering short audio clips of her voice and her visual likeness.

As a law professor, I was struck by Swift’s filings because they highlight a new legal frontier in artificial intelligence.

Most AI-related litigation has centered on copyright law, which protects creative works such as songs, books, photographs and recordings from being copied, distributed, adapted or publicly performed without permission.

But TAS Rights Management’s recent move involves trademark law, not copyright. The filings aren’t really about protecting Swift’s lyrics or albums. Instead, they’re about preventing AI-generated voices and images from misleading people into believing she has endorsed a product, political message or cause.

Copyright is about creative works

Most AI-related lawsuits have been tied to whether copyright violations have taken place – specifically, whether AI companies used copyrighted works to train their systems, or whether their chatbots have produced outputs that too closely resemble protected material.

For example, The New York Times sued OpenAI and Microsoft in 2023, alleging that the companies used the outlet’s journalism to train their AI systems, which then went on to generate outputs that have competed with or reproduced New York Times articles. Authors, publishers, photo agencies and music publishers have sued other AI companies for the same reason.

But copyright violations are only one part of the legal issues raised by generative AI.

Copyright doesn’t necessarily protect a person’s identity. It does not give Swift a general right to control anything that sounds like her, looks like her or evokes her in the minds of audiences.

If an AI-generated voice imitates Swift without copying a particular recording, song or lyric, copyright may not address the real issue, which is that people are being led to believe she said, sang or endorsed something she never approved.

Trademarks are about trust

Trademark law starts from a different concern. It protects names, images, sounds and other markers that help consumers identify who or what is behind a product or service.

A trademark can be a word, phrase, symbol, design or combination of these things. Familiar examples include brand names such as Coca-Cola, logos like the Nike swoosh, slogans like Subway’s “Eat Fresh” and even distinctive sounds, such as the MGM lion roar.

A red banner featuring the Coca-Cola logo and the text 'FIFA World Cup 26.'
FIFA uses a ‘TM’ wordmark in its 2026 World Cup logo, meaning soccer’s world governing body is claiming the logo as a trademark. Coca-Cola features a small ‘R’ with a circle around it at the end of its iconic cursive logo to indicate that it has registered the design as a trademark with the United States Patent and Trademark Office. Steve Russell/Toronto Star via Getty Images

A trademark is not a general ownership right over a word, phrase, voice or image. It is a way of helping consumers know who stands behind what they are buying, hearing or seeing.

That difference becomes crucial once AI can mimic a person’s voice or face. Suppose a company uses an AI-generated Swift-like voice to sell perfume or cryptocurrency. The concern is that listeners may think Swift approved of the product or message.

That is a trademark problem. Trademark law asks whether the use misleads consumers about whether a company or person has produced or endorsed something. Swift’s filings appear aimed at that danger. They suggest a concern beyond copied songs: fake endorsements, fake appearances and fake signals of approval.

Swift’s concerns also bleed into what are known as “publicity rights,” which generally protect against unauthorized commercial use of a person’s identity, such as a name, image, likeness or voice.

A classic publicity rights case involves a company using a celebrity’s face in an advertisement without permission to mislead consumers into believing the celebrity endorses the product.

AI’s ability to clone voices and images makes publicity law especially relevant. But in the United States, publicity rights are mostly governed by state law, and the rules vary widely from one state to another. That patchwork helped inspire the bipartisan NO FAKES Act, introduced in 2025, which would create a national standard that would prohibit unauthorized AI-generated replicas of a person’s voice or visual likeness. The bill, still in its early stages, has been referred to the Senate Judiciary Committee for consideration.

The untested part

Swift is not alone: Actor Matthew McConaughey trademarked “alright alright alright,” his memorable line from “Dazed and Confused,” to protect it from being used in AI-generated content.

The courts have already affirmed that sounds can function as trademarks. But it isn’t clear whether trademark law can police AI-generated replicas of a person’s voice or image when the issue is not counterfeiting but a manufactured endorsement.

A person’s voice or likeness is not automatically a trademark. In order to qualify as one, it must be used help consumers identify who is behind a product or service.

One existing limit on trademark protection is especially important. Federal law protects certain uses of a celebrity’s image and likeness in cases involving parody, criticism, commentary and news reporting. Not every imitation is a form of deception.

Courts will have to draw that line on a case-by-case basis. A fake ad that makes consumers think Swift endorsed a product is different from a parody that comments on celebrity culture. A scam using her voice is different from a news story about AI deepfakes.

That said, Swift’s filings reflect a real problem: AI has allowed fake endorsements to look and sound real enough to spread before anyone has time to set the record straight.

Major AI copyright cases will continue to focus on copied works. But when AI is used to manufacture identity, endorsement or trust, copyright alone is no longer enough. Swift’s filings suggest that AI law will increasingly focus not only on protecting the work of musicians, writers, journalists and artists, but also on protecting the signals that tell audiences who is really speaking.

Daryl Lim, Associate Dean for Research and Strategic Partnerships, Penn State

This article is republished from The Conversation under a Creative Commons license. Read the original article.

The Conversation