Policies > Artificial Intelligence > AI and Elections

Artificial Intelligence
& Elections

For all its upside potential, generative AI has the ability to drastically increase the spread of election mis- and disinformation and cause confusion among voters.

AI is already playing a role in the 2024 election. In the New Hampshire primary, a robocall using AI technology to impersonate President Biden sought to mislead voters, discouraging them from voting. Following this activity, the Federal Communications Commission issued a declaratory ruling that restricted robocalls containing voices generated by AI.

Much more needs to be done. States and localities are on the front lines when it comes to elections, and their elected leaders must address concerns about this growing technology.

As Minnesota’s Secretary of State Steve Simon recently said at a NewDEAL Forum convening on the subject, voter suppression in America is not new. In today’s environment, generative AI has the potential to amplify existing threats, and officials must adapt their efforts to combat misinformation. The how changes, but the what remains the same.

In order to assist state and local elected officials in preparing for the 2024 election, the NewDEAL Forum compiled five best practices for mitigating the negative impacts of AI in the near term, as well as principles for AI-focused legislation to protect our democracy.

Addressing Generative AI in the 2024 Election

Principles for AI-focused Legislation to Protect Democracy

The NewDEAL Forum has identified six broad principles that should be incorporated into any legislation regulating generative-AI and campaigning. These recommendations follow a review of dozens of proposed and enacted pieces of legislation as well as discussions with issue-area experts. Any policy or legislation dealing with AI and elections should, at a minimum, include the following six components:

Longer Term Considerations

Issue-area experts and elected officials partnering with the NewDEAL Forum have agreed on three additional areas that are still critical as AI technology advances:

  • Verification

    Policymakers should support ways to verify content as quickly as possible. This could be a “digital marker” of some sort that is indelible for official communications, in order to give people confidence in their trustworthiness

  • Funding

    There is broad consensus that more funding is needed for state and local election officials to administer safe, secure, and fair elections. In particular, more funding is needed to counter AI-generated cyber threats aimed at elections and election officials, as well as additional technical assistance.

  • Adaptation

    Because of the dynamic nature of generative AI, and its exponential growth potential, policymakers will need a mechanism to regularly re-evaluate laws and policies in light of advancements. Regulations should both protect voters and citizens as well as ensure AI can achieve beneficial applications.

What’s next? See more Policies from the NewDeal Forum