Just days after sexually explicit images of artificial intelligence-powered Taylor Swift went viral on X, searches for her name on the platform formerly known as Twitter are no longer returning results.
As of Monday morning, all searches for Taylor Swift are displaying the message: “An error has occurred. Try reloading.” However, if her name is enclosed in quotation marks, posts mentioning her name may appear.
The search error came after a series of sexually explicit deepfake images of Swift circulated on the social media site, angering fans and highlighting harmful effects of the technology.
Joe Benarroch, Head of Business Operations at
USA TODAY has reached out to Swift's rep for comment.
In a mock photo created using AI-powered image generators, Swift is seen posing inappropriately at a Kansas City Chiefs game. The Grammy winner has been increasingly seen in real life at the team's games, supporting football star Travis Kelce.
Following the backlash to the images, X released a statement regarding his security account.
“Posting images containing non-consensual nudity (NCN) is strictly prohibited on X and we have a zero-tolerance policy towards such content,” the post reads. “Our teams are actively removing all identified images and taking appropriate action against the accounts responsible for publishing these images.”
Swift has not publicly commented on the images.
The singer was still smiling as she attended the Chiefs' game against the Baltimore Ravens on Sunday and gave Kelce a kiss. The Chiefs are now headed to the Super Bowl after defeating the Ravens 17-10.
What you need to know:Sexually explicit AI images of Taylor Swift are circulating online, sparking backlash
Deepfake AI images and videos of celebrities increasingly seen online
AI images can be created using text prompts and generated without the subject's consent, raising privacy concerns.
AI-generated deepfakes – manipulated videos created using machine learning techniques to produce realistic but fake images and sound – are also increasingly being used to create fake celebrity endorsements.
A variety of other fake images have spread online in recent years, including photos showing former President Donald Trump being arrested, assaulted and carried away by a group of police officers that went viral on social media last year. Currently, it is still possible to look closely at AI-generated images and find clues that they are not real. For example, one of the images from Trump's arrest showed him with three legs.
But experts say it's only a matter of time before there's no way to visually distinguish between a real image and an AI-generated image.
“I'm very confident that in the long run it will be impossible to tell the difference between a generated image and a real one,” James O'Brien, a computer science professor at the University of California, Berkeley, told USA TODAY. “The images generated are getting better and better.”
Artificial intelligence:AI-generated explicit photos of Taylor Swift are just the tip of the iceberg for the threat of deepfakes
Contributors: Chris Mueller, USA TODAY; The Associated Press
Comments are closed.