House Bill Proposes Mandatory Disclosure of AI-Generated Content in Political Advertisements

In response to growing concerns over AI-generated misinformation, Representative Yvette Clarke, a Democrat from New York, has proposed the REAL Political Ads Act. This legislation aims to enhance transparency in political advertising by requiring that any use of generative AI be clearly disclosed via prominent audio or text notifications. The act serves as an amendment to the Federal Election Campaign Act and mandates the Federal Election Commission (FEC) to establish regulations for enforcement. Notably, the law is set to take effect on January 1, 2024, regardless of whether these regulations are established.

Clarke emphasizes the urgent need for this measure ahead of the 2024 election cycle, citing the potential of generative AI to "manipulate and deceive people on a large scale." She warns that unchecked AI usage could have severe repercussions for elections and national security, stating that existing laws have not kept pace with technological advancements.

This proposal follows a recent incident where Republicans deployed AI-generated visuals in a political ad, suggesting potential scenarios during President Biden's hypothetical second term. While the ad includes a subtle disclaimer noting its AI-generated nature, there are concerns that future ads may bypass such disclaimers or present falsehoods as fact.

The push for AI regulation is not isolated. California Representative Ted Lieu has introduced broader measures for AI oversight, while the National Telecommunications and Information Administration (NTIA) is seeking public input on potential AI accountability frameworks. In contrast, Clarke’s legislation is specifically focused on the realm of political advertising, aiming for swift passage.

The path to enactment remains uncertain. The bill must navigate a vote in the Republican-controlled House and find parallel support in the Senate before Congress can reconcile the two versions and present a unified law to the President. Nonetheless, even if passed, the law may not eliminate all attempts to mislead voters. However, it could serve as a deterrent, encouraging political entities to use AI responsibly and ethically.