With the expansion of artificial intelligence comes politicians using it for political gain – sometimes deceitfully.

This summer, the campaign for Florida Gov. Ron Desantis was criticized for posting AI-generated deep fakes of former President Donald Trump and Dr. Anthony Fauci embracing, leading to much criticism from other Republicans.

In Pennsylvania, a bipartisan team is trying to tackle the use of deepfakes in state elections.

The legislation, led by Rep. Tarik Khan, D-Philadelphia, would fine candidates, campaigns and Super PACs if they use deep fake technology to fraudulently misrepresent other candidates. The fine amounts are still being finalized, said Khan.

“The idea of this legislation is to really start to put the guardrails on the process and to make sure that our elections remain safe and secure,” said Khan.

Co-sponsor Rep. Rob Mercuri, R-Allegheny, shares the same sentiment.

“There’s no doubt that voters deserve truthful, accurate information,” Mercuri said. “And so creating guardrails where they know that candidates and other campaign entities such as Super PACs would be prohibited from using and misusing candidate images and video.

Mercuri said this bill is important for future elections.

“I want to ensure a safe and effective elections process that presents trusted information to the voters so that they can make an informed choice,” he said.

If moved, Pennsylvania would become one of a handful of states with laws on the book dealing with false, AI-generated images in elections.

Khan said the advent of AI can lead people to be duped by false information.

“People are good. They’re trusting,” he said. “And all too often people are victims of scams.”

The law would only affect campaigns, candidates and super PACs, so those popular videos of presidents playing video games would not be affected nor would late night shows that have been making satirical images of candidates.

“We are not intending here to take away the ability of someone to use their free speech,” said Mercuri.

In a 2022 post related to deep fake laws, conservative think tank Institute for Free Speech argued that “government attempts to outlaw misleading political messages will do far more harm to democracy than the mere existence of such speech.”

The group argues voters can “expose falsehoods and decide the value of political expression themselves.”

The Federal Election Commission is considering regulating AI deepfakes in campaign ads if used to misrepresent political opponents.

Additionally, X, formerly known as Twitter, and Meta, the parent company of Facebook, have rules regarding the use of deep fakes.

The Shapiro administration will continue reviewing the legislation as it moves through the process, according to press secretary Manuel Bonder.