Legislation would require clear disclosure when generative AI is used to provide election-related information to voters
(TRENTON) — In response to growing concerns over the role artificial intelligence (AI) may play in spreading misinformation during elections, the Assembly State and Local Government Committee today advanced bill A4729, which would require a disclosure when generative AI is used to operate chatbots providing election-related information to voters. Sponsored by Assembly members Robert Karabinchack, Ellen Park, and Avi Schnall, the legislation seeks to increase transparency for voters and help voters better identify when information is being produced by AI rather than a human.
“As artificial intelligence becomes more integrated into our everyday lives, more people are turning to these systems with the belief that the information they receive is accurate and trustworthy,” said Assemblyman Karabinchack, Chair of the Assembly State and Local Government Committee (D-Middlesex). “When it comes to the sanctity of our elections, voters deserve to know if the information they are receiving is coming from a human being or being generated by a computer.”
Under the bill, any AI chatbot using generative AI to create audio, video, text, or print content for the purpose of providing voters with election-related information or information concerning the accomplishments, policy positions, or qualifications of a candidate for election in New Jersey would be required to clearly disclose that the content is being generated by an AI system.
“New Jersey proudly stands at the forefront of the rapidly evolving innovation economy and the unique opportunities created by artificial intelligence,” said Assemblywoman Park (D-Bergen). “But with those opportunities also comes a responsibility to address the risks that can arise when misleading information is given the room to gain traction before it can be corrected. As more people inevitably rely on AI-generated content to make decisions, ensuring transparency around where that information comes from will be so important.”
A 2025 study published in the Harvard Misinformation Review found that 83 percent of Americans are concerned about AI being used to spread election misinformation, while researchers have also warned that AI systems can generate inaccurate polling locations, voter deadlines, and ballot procedures that may mislead voters.
“This legislation takes a practical and responsible approach to making sure our voters have transparency when they go seeking information that could influence our democratic process,” said Assemblyman Schnall (D-Monmouth, Ocean). “People should not have to guess whether the information they are receiving is coming from a real person and this bill does exactly that at a time when that trust matters more than ever.”