Former President Donald Trump took to posting what seemed to be a sequence of AI-generated images during the weekend in an attempt to garner support for his presidential candidacy. Among these, there was a falsely purported endorsement from the pop star Taylor Swift.

These posts illustrate how Trump might utilize generative AI in a manner that complicates efforts to regulate AI-created election misinformation. This is partly due to the long-standing legal precedent that permits candidates to deceive in political advertisements. They came shortly after Trump wrongly accused his opponent, Vice President Kamala Harris, of using AI to generate a rally crowd.

The images Trump shared included one that seemingly depicted Harris from the back as she addressed a crowd in Chicago, the location of the Democratic National Convention this week, with a communist hammer and sickle prominently featured in the background.

Another post incorporated screenshots of other users’ posts showing images of “Swifties for Trump,” along with what appeared most evidently to be an AI-generated image of Taylor Swift dressed as Uncle Sam and bearing the words, “Taylor wants you to vote for Donald Trump.” Alongside the collection of screenshots (only one of which was marked as satire), Trump wrote, “I accept!”

Trump’s posts are likely not covered by the increasing number of state laws against election deepfakes, as stated by Robert Weissman, the co-president of Public Citizen. While approximately 20 states have implemented regulations regarding the use of AI-generated false images in elections, they typically prohibit depictions of someone doing or saying something in a persuasive manner. “So, it’s not just a well-executed deepfake or a proficient generative AI application, but it actually has to be plausible,” Weissman remarks.

Meanwhile, when it comes to election misinformation, “there are no federal restrictions on the use of deepfakes,” Weissman says, noting the exception of the Federal Communications Commission’s ban on robocalls that employ AI-generated voices. The non-profit consumer advocacy group has been striving to prompt the Federal Election Commission to limit candidates’ ability to misrepresent their opponents using AI, but its rules likely wouldn’t cover something as overtly exaggerated as the Harris image or something that doesn’t depict an opposing candidate, as in the case of the Swift image.

Still, Weissman suggested that Swift might have a claim for the use of her likeness to falsely issue an endorsement, perhaps under California’s Right of Publicity, which safeguards the use of a person’s likeness. Universal Music Group, representing Swift, did not promptly respond to a request for comment regarding the use of her likeness in Trump’s post. The Trump campaign also failed to respond immediately.

Courts have consistently determined that the First Amendment often shields even intentional falsehoods, including those made by political candidates. Even if Congress were to pass regulations concerning AI deepfakes, it wouldn’t halt many of their uses. Weissman says an aggrieved opponent would need to demonstrate that a lie resulted in harm or injury to voters, for instance, to render it illegal. “I don’t think that our legislative solutions are going to be flawless, even if we obtain exactly what we desired,” he states.

Private platforms could take measures against misleading generative AI content, of course, without government intervention. X’s synthetic and manipulated media policy prohibits such posts “that may deceive or confuse people and lead to harm.” But the platform has seemingly enforced the rule selectively, with even its owner, Elon Musk, seeming to violate it by posting a deepfake of Harris not clearly marked as parody. Trump’s other preferred venue, his own platform Truth Social, has minimal rules in its community guidelines.

“It’s convenient for Trump, who was previously dismissing everything as fake before the advent of AI and wants us to consider true things as fake – like Harris’ crowds – to disseminate AI-generated nonsense to undermine the very concept of authenticity and, to some extent, reality,” says Weissman. “It’s extremely challenging to have a democratic society if people can’t trust what they see and hear with their own eyes.”