Taylor Swift’s Instagram post endorsing Vice President Kamala Harris for president, and citing AI-generated deepfake images of Swift posted by Donald Trump, led about 406,000 people to click on a voter-registration information site, The New York Times reported on Thursday. Swift’s post linked to Vote.gov, a site that directs users to their own state’s voter-registration site, so it’s unclear how many of the people who clicked the link registered in their state.
The Times notes that Swift’s link was responsible for more than half of the roughly 727,000 visitors the site received from Tuesday to Wednesday.
The popular singer-songwriter, who had previously been silent about her stance on the candidates, announced her endorsement after Tuesday night’s debate between Harris and Trump, saying that the deepfakes Trump promoted “conjured up my fears around AI, and the dangers of spreading misinformation.”
“It brought me to the conclusion that I need to be very transparent about my actual plans for this election as a voter,” Swift said. “The simplest way to combat misinformation is with the truth.”
Swift’s endorsement of the Democratic nominee for president comes just weeks after Trump posted on his Truth Social platform that he had accepted Swift’s support. He also posted AI-generated deepfake images of Swift and her fans, known as Swifties, seemingly showing support for the Republican presidential nominee.
The images, which Trump topped with the text “I accept,” were originally posted on X, formerly known as Twitter, by a user who labeled them as satire. One of the images reposted on Trump’s Truth Social account had the word “satire” in the image text.
Conversely, Trump falsely accused the Harris campaign of using AI to fake a photo taken at a rally in August to show a greatly inflated crowd size. But numerous other videos and photos of the event showed a crowd similar in size to the one shown in the Harris campaign photo.
Election Deepfakes Are Here and Better Than Ever
The advancement of artificial intelligence, most notably with the rapid evolution of generative AI tools over the past two years, has made it easier for people to generate convincing deepfakes — manipulated images, video or audio appearing to show people doing or saying things they didn’t actually say or do. Experts worry these deepfakes could affect how or even if people vote ahead of the November presidential election.
The best way to defend against deepfakes, experts say, is to be skeptical and to have an eye for detail. AI-generated deepfake videos, for instance, may have quirks in motion, lighting or phrasing. It’s also important to evaluate whether the source is reputable and to do some fact-checking of your own.
The Trump campaign didn’t respond to a request for comment.
Watch this: Deepfake Tech at Defcon 2024
Read the full article here
