![FILE - People vote on Election Day, Nov. 7, 2023, at Knox Presbyterian Church in Cincinnati. On Monday, January 8, 2024, a federal judge upheld as constitutional provisions of a comprehensive election law that Ohio enacted last year, related to strict new photo ID requirements, drop box restrictions, and absenteeism. The Democratic Party law firm's challenge to the stricter deadline for and provisional ballots. (AP Photo/Joshua A. Bickel, File)](https://www.kxan.com/wp-content/uploads/sites/40/2024/01/659db51bce82b9.12981623.jpeg?w=900)
FILE – People vote on Election Day, Nov. 7, 2023, at Knox Presbyterian Church in Cincinnati. On Monday, January 8, 2024, a federal judge upheld as constitutional provisions of a comprehensive election law that Ohio enacted last year, related to strict new photo ID requirements, drop box restrictions, and absenteeism. The Democratic Party law firm’s challenge to the stricter deadline for and provisional ballots. (AP Photo/Joshua A. Bickel, File)
AUSTIN (NEXSTAR) – With less than a month until early voting in the Texas primary, technology experts warn about the role of artificial intelligence in election misinformation following several high-profile incidents involving AI. are doing.
Reports of robocalls using AI to imitate President Joe Biden’s voice surfaced two weeks ago, threatening voters as they headed to the polls ahead of New Hampshire’s Jan. 23 primary. It seems to have dissuaded him. The state attorney general then launched an investigation, saying the call appeared to be an illegal attempt at voter suppression.
Further questions are arising about how AI will affect the 2024 election, with state legislatures across the country proposing bans or mandatory disclosure of AI-generated election content.
Jelly Martin, a graduate research assistant in the Propaganda Lab at the University of Texas at Austin, said recent AI-powered misinformation campaigns highlight the growing need for technology regulation. Ta.
“We need to think critically about these concerns without overhyping or panicking about the new technologies that are coming,” she said. “If you don’t limit the technology, it can become very convincing very quickly.”
The federal Cybersecurity and Infrastructure Security Agency (CISA) released a study describing the known capabilities and uses of generative AI. The list included a wide range of malicious activity, from deepfake videos of news anchors to fake accounts used in phishing campaigns in online chat rooms.
Additionally, AI-generated photos could be used to create fake social media accounts that look realistic and trustworthy.
CISA also flagged potential concerns directly related to the election. AI capabilities provide leverage both domestically and internationally to increase the scale and penetration of disinformation campaigns.
“If you have deepfakes and videos of Mr. Biden saying things just like him that he didn’t actually say, people are going to have a very hard time parsing fact from fiction,” Martin said. Ta. “Please double check your sources. Look for reliable sources.”
According to the study, AI could be used to generate malware that evades detection systems and creates “convincingly false election records.” Cloning technology could be used to imitate the voices of election officials and gain access to sensitive information.
Challenges in regulating AI
In 2019, Texas became the first state to ban the creation and distribution of deepfake videos intended to influence elections or harm candidates with Senate Bill 751.
The law criminalizes videos released within 30 days before an election. There is no mention of deepfake photos or audio, which is the only section of the Texas election law related to AI.
Texas established an artificial intelligence advisory committee last June. The council is tasked with overseeing how the technology is used in state government agencies and recommending legislative changes. Government agencies are required to report information to the board by July 2024.
Congress has increasingly held hearings and debates about AI, but has failed to pass legislation that would limit the use of the technology or add penalties for harmful uses.
In 2023, six bills addressing deepfakes in campaigns and elections were introduced on Capitol Hill, two of which passed the House, but none ultimately made it to the president’s desk.
Martin expressed doubts that U.S. lawmakers will be able to pass meaningful legislation to help prevent AI-driven misinformation and disinformation, as Congress has sought to better regulate social media since its inception.
“Regulation is so slow because we weren’t proactive when it happened,” she said. “Policymakers may not agree on the best way to do it quickly enough, so we’re kind of at a disadvantage there…While it’s not a perfect solution, The best solution is probably still education.”