How AI is reshaping the 2024 presidential election

The dawn of AI in elections: The 2024 U.S. election marks a significant milestone as the first where generative AI tools are widely accessible to the public, political actors, and foreign agents, potentially impacting the democratic process.

• Alexandra Reeve Givens, president and CEO of the Center for Democracy and Technology (CDT), describes generative AI as a “magnifier to existing influence efforts” that can make disinformation cheaper, easier, and potentially more convincing.
• The widespread availability of AI tools has changed how voters access election information and has increased the volume and sophistication of misinformation.

AI chatbots and voter misinformation: Concerns have been raised about the reliability of AI chatbots as sources of election information, with studies showing significant inaccuracies in their responses.

• The AI Democracy Projects’ report found that half of the responses from five leading AI models to election-related queries were rated as inaccurate by expert testers.
• More than one-third of responses were rated as harmful or incomplete, with an additional 13 percent rated as biased.
• A Reuters Institute analysis of chatbot responses to European Parliament election questions also revealed instances of partially correct or false and misleading information.

Foreign interference amplified by AI: The Office of the Director of National Intelligence (ODNI) has reported that foreign actors, particularly Russia and Iran, are leveraging AI to interfere with U.S. elections.

• Russia has generated the most AI content related to the election across text, images, audio, and video mediums.
• AI-generated disinformation is being used to boost certain candidates, denigrate others, and amplify divisive issues such as immigration.
• A recent Russian influence campaign attempted to mislead people about Haitian immigrants casting illegal votes in Georgia, exploiting anti-immigrant sentiment and concerns about voter fraud.

Combating AI-driven misinformation: Various strategies are being employed to address the challenges posed by AI-generated misinformation in elections.

• Election officials are working to boost trusted information sources for voters.
• Secretaries of State are quickly debunking deepfake videos and images circulated by foreign actors.
• “Prebunking” campaigns are gaining traction as a method to inoculate voters against misinformation by exposing them to weakened doses of false information paired with explanations.

Regulatory responses and tech industry responsibility: Efforts are being made at both governmental and industry levels to address the challenges posed by AI in elections.

• The Biden Administration’s AI Executive Order includes provisions to develop authentication techniques for government communications.
• The CDT has urged the Federal Election Commission to address the use of misleading deepfake images by political campaigns.
• Tech companies are being called upon to be transparent about the limitations of their AI tools and to direct users to authoritative sources of election information.

AI in political campaigns: Recent research reveals complex attitudes among U.S. voters towards the use of AI in political campaigns.

• While voters are generally wary of AI in relation to political campaigns, they show particular concern about its potential for deception.
• Interestingly, political parties may face few consequences from supporters for using AI deceptively in campaigns.
• Voters show more acceptance of AI being used for basic campaign operations like content generation, indicating they can distinguish between legitimate and concerning applications of the technology.

Broader implications: The integration of AI into the electoral process raises significant questions about the future of democratic discourse and the integrity of elections.

• The ability of AI to amplify existing influence efforts and create more convincing disinformation poses a serious challenge to election integrity and voter trust.
• The acceptance of deceptive AI practices by partisan supporters highlights the potential for technology to further polarize the electorate.
• As AI continues to evolve, it will be crucial for policymakers, tech companies, and voters to adapt and develop new strategies to ensure the reliability of election information and maintain the integrity of the democratic process.