Republican uses AI to write legislation regulating deep fakes

An Arizona Republican got caught using the artificial intelligence site ChatGPT to craft legislation about the technology used to craft so-called deepfake videos, NBC News reported.

State Rep. Alex Kolodin needed help crafting the definition of “digital impersonation” for a new law that would regulate false videos of people saying or doing things that never occurred.

“I used it to write the part of the bill that had to do with defining what a deepfake was,” Kolodin said of the move. “I was really struggling with the technical aspects of how to define what a deepfake was. So I thought to myself, Well, why not ask the subject matter expert, ChatGPT?"

The legislation in question, House Bill 2394, was signed by Democratic Gov. Katie Hobbs after being passed by both chambers of the state legislature without being broadly noticed.

Read Also: We asked a government agency about AI. They sent us 62 blacked-out pages.

Typically, laws are drafted by lawmakers with help from legislative counsel. Kolodin didn't explain why he was writing solo.

He explained that he “uploaded the draft of the bill that I was working on and said, you know, please, please put a subparagraph in with that definition, and it spit out a subparagraph of that definition.”

Unlike others who have used AI to do things, Kolodin said that he read through the results to ensure accuracy.

"If ChatGPT had effed up some of the language or did something that would have been harmful, I would have spotted it, one of the 10 stakeholder groups that worked on or looked at this bill, the ACLU would have spotted, the broadcasters association would have spotted it, it would have got brought out in committee testimony," he said.

One of the more recent AI flubs occurred when former Donald Trump lawyer Michael Cohen used a legal AI program to search for case law on an issue. That site, powered by Google, gave false examples of cases that never existed. Once Cohen discovered the error, he told the court and replaced the citations with legitimate ones.

Read the full report here.