Taylor Swift’s affinity for Le Creuset is real: Her collection of the cookware has been featured on a Tumblr account dedicated to the pop star’s home décor, showcased in her gift choices at a fan’s bridal shower and shown in a Netflix documentary that was highlighted by Le Creuset’s Facebook page.
What is not real: Ms. Swift’s endorsement of the company’s products, which have appeared in recent weeks in ads on Facebook and elsewhere featuring her face and voice.
The ads are among the many celebrity-focused scams made far more convincing by artificial intelligence. Within a single week in October, the actor Tom Hanks, the journalist Gayle King and the YouTube personality MrBeast all said that A.I. versions of themselves had been used, without permission, for deceptive dental plan promotions, iPhone giveaway offers and other ads.
In Ms. Swift’s case, experts said, artificial intelligence technology helped create a synthetic version of the singer’s voice, which was cobbled together with footage of her alongside clips showing Le Creuset Dutch ovens. In several ads, Ms. Swift’s cloned voice addressed “Swifties” — her fans — and said she was “thrilled” to be handing out free cookware sets. All people had to do was click on a button and answer a few questions before the end of the day.
Le Creuset said it was not involved with the singer for any consumer giveaway. The company urged shoppers to check its official online accounts before clicking on suspicious ads. Representatives of Ms. Swift, who was named Person of the Year by Time magazine in 2023, did not respond to requests for comment.
Famous people have lent their celebrity to advertisers for as long as advertising has existed. Sometimes, it has been unwillingly. More than three decades ago, Tom Waits sued Frito-Lay — and won nearly $2.5 million — after the corn chip company imitated the singer in a radio ad without his permission. The Le Creuset scam campaign also featured fabricated versions of Martha Stewart and Oprah Winfrey, who in 2022 posted an exasperated video about the prevalence of fake social media ads, emails and websites falsely claiming that she endorsed weight loss gummies.
Over the past year, major advances in artificial intelligence have made it far easier to produce an unauthorized digital replica of a real person. Audio spoofs have been especially easy to produce and difficult to identify, said Siwei Lyu, a computer science professor who runs the Media Forensic Lab at the University at Buffalo.
The Le Creuset scam campaign was probably created using a text-to-speech service, Dr. Lyu said. Such tools usually translate a script into an A.I.-generated voice, which can then be incorporated into existing video footage using lip-syncing programs.
“These tools are becoming very accessible these days,” said Dr. Lyu, who added that it was possible to make a “decent-quality video” in less than 45 minutes. “It’s becoming very easy, and that’s why we’re seeing more.”
Dozens of separate but similar Le Creuset scam ads featuring Ms. Swift — many of them posted this month — were visible as of late last week on Meta’s public Ad Library. (The company owns Facebook and Instagram.) The campaign also ran on TikTok.
The ads sent viewers to websites that mimicked legitimate outlets like the Food Network, which showcased fake news coverage of the Le Creuset offer alongside testimonials from fabricated customers. Participants were asked to pay a “small shipping fee of $9.96” for the cookware. Those who complied faced hidden monthly charges without ever receiving the promised cookware.
Some of the fake Le Creuset ads, such as one mimicking the interior designer Joanna Gaines, had a deceptive sheen of legitimacy on social media thanks to labels identifying them as sponsored posts or as originating from verified accounts.
In April, the Better Business Bureau warned consumers that fake celebrity scams made with A.I. were “more convincing than ever.” Victims were often left with higher-than-expected charges and no sign of the product they had ordered. Bankers have also reported attempts by swindlers to use voice deepfakes, or synthetic replicas of real people’s voices, to commit financial fraud.
In the past year, several well-known people have publicly distanced themselves from ads featuring their A.I.-manipulated likeness or voice.
This summer, fake ads spread online that purported to show the country singer Luke Combs promoting weight loss gummies recommended to him by the fellow country musician Lainey Wilson. Ms. Wilson posted an Instagram video denouncing the ads, saying that “people will do whatever to make a dollar, even if it is lies.” Mr. Combs’s manager, Chris Kappy, also posted an Instagram video denying involvement in the gummy campaign and accusing foreign companies of using artificial intelligence to replicate Mr. Combs’s likeness.
“To other managers out there, A.I. is a scary thing and they’re using it against us,” he wrote.
A TikTok spokesperson said the app’s ads policy requires advertisers to obtain consent for “any synthetic media which contains a public figure,” adding that TikTok’s community standards require creators to disclose “synthetic or manipulated media showing realistic scenes.”
Meta said it took action on the ads that violated its policies, which prohibit content that uses public figures in a deceptive way to try to cheat users out of money. The company said it had taken legal steps against some perpetrators of such schemes, but added that malicious ads were often able to evade Meta’s review systems by cloaking their content.
With no federal laws in place to address A.I. scams, lawmakers have proposed legislation that would aim to limit their damage. Two bills introduced in Congress last year — the Deepfakes Accountability Act in the House and the No Fakes Act in the Senate — would require guardrails such as content labels or permission to use someone’s voice or image.
At least nine states, including California, Virginia, Florida and Hawaii, have laws regulating A.I.-generated content.
For now, Ms. Swift will probably continue to be a popular subject of A.I. experimentation. Synthetic versions of her voice pop up regularly on TikTok, performing songs she never sang, colorfully sounding off on critics and serving as phone ringtones. An English-language interview she gave in 2021 on “Late Night With Seth Meyers” was dubbed with an artificial rendering of her voice speaking Mandarin. One website charges up to $20 for personalized voice messages from “the A.I. clone of Taylor Swift,” promising “that the voice you hear is indistinguishable from the real thing.”
