State lawmakers eye promise, pitfalls of AI ahead of November elections
Sep. 12—LOUISVILLE, Ky. — Inside a white-walled conference room, a speaker surveyed hundreds of state lawmakers and policy influencers, asking whether artificial intelligence poses a threat to the elections in their states.
The results were unambiguous: 80% of those who answered a live poll said yes. In a follow-up question, nearly 90% said their state laws weren't adequate to deter those threats.
It was among the many exchanges on artificial intelligence that dominated sessions at last month's meeting of the National Conference of State Legislatures, the largest annual gathering of lawmakers, in Louisville.
"It's the topic du jour," Kentucky state Sen. Whitney Westerfield, a Republican, told lawmakers as he kicked off one of many panels centering on AI. "There are a lot of discussions happening in all of our state legislatures across the country."
While some experts and lawmakers celebrated the promise of AI to advance services in health care and education, others lamented its potential to disrupt the democratic process with just months to go before November's elections. And lawmakers compared the many types of legislation they're proposing to tackle the issue.
This presidential election cycle is the first since generative AI — a form of artificial intelligence that can create new images, audio and video — became widely available. That's raised alarms over deepfakes, remarkably convincing but fake videos or images that can portray anyone, including candidates, in situations that didn't occur or saying things they didn't.
"We need to do something to make sure the voters understand what they're doing," said Kentucky state Sen. Amanda Mays Bledsoe.
The Republican lawmaker, who chairs a special legislative task force on AI, co-sponsored a bipartisan bill this year aimed at limiting the use of deepfakes to influence elections. The bill would have allowed candidates whose appearance, action or speech was altered through "synthetic media" in an election communication to take its sponsor to court. The state Senate unanimously approved the proposal but it stalled in the House.
While Bledsoe expects to bring the bill up again next session, she acknowledged how complex the issue is: Lawmakers are trying to balance the risks of the evolving technology against their desire to promote innovation and protect free speech.
"You don't want to go too fast," she said in an interview, "but you also don't want to be too behind."
Rhode Island state Sen. Dawn Euer, a Democrat, told Stateline she's concerned about AI's potential to amplify disinformation, particularly across social media.
"Election propaganda and disinformation has been part of the zeitgeist for the existence of humanity," said Euer, who chairs the Senate Judiciary Committee. "Now, we have high-tech tools to do it."
Connecticut state Sen. James Maroney, a Democrat, agreed that concerns about AI's effects on elections are legitimate. But he emphasized that most deepfakes target women with digitally generated nonconsensual intimate images or revenge porn. Research firm Sensity AI has tracked online deepfake videos for years, finding 90% of them are nonconsensual porn, mostly targeting women.
Maroney sponsored legislation this year that would have regulated artificial intelligence and criminalized deepfake porn and false political messaging. That bill passed the state Senate, but not the House. Democratic Gov. Ned Lamont opposed the measure, saying it was premature and potentially harmful to the state's technology industry.
While Maroney has concerns about AI, he said the upsides far outweigh the risks. For example, AI can help lawmakers communicate with constituents through chatbots or translate messaging into other languages.
Top election officials on AI
During one session in Louisville, New Hampshire Republican Secretary of State David Scanlan said AI could improve election administration by making it easier to organize election statistics or get official messaging out to the public.
Still, New Hampshire experienced firsthand some of the downside of the new technology earlier this year when voters received robocalls that used artificial intelligence to imitate President Joe Biden's voice to discourage participation in a January primary.
