LobbyComply Blog

The U.S. District Court for the District of Hawaii recently ruled that a 2024 state law (Act 191) aimed at regulating deceptive election media is unconstitutional. In the case of The Babylon Bee, LLC and Dawn O’Brien v. Anne E. Lopez, et al., the court granted a permanent injunction against the law, finding that it violated the First and Fourteenth Amendments. While the state argued the law was necessary to protect electoral integrity from AI-generated “deepfakes,” the court determined the restrictions were too broad and lacked the narrow tailoring required by the Constitution.

Understanding Act 191 and the Court’s Ruling

Act 191 was designed to prevent the reckless distribution of “materially deceptive” election media. This included any video, audio, or images created via digital technology or artificial intelligence that depicted a candidate engaging in speech or conduct that did not actually happen. The law specifically targeted content intended to harm a candidate’s reputation or influence voter behavior from February through Election Day.

The court’s decision centered on several key legal principles:

• Content and Speaker Discrimination: The court found that Act 191 discriminated based on the content of the speech and the identity of the speaker.
• Failure of Strict Scrutiny: Even though protecting elections from deceptive media is a “compelling interest,” the law failed the strict scrutiny test because it was not narrowly tailored.
• Vagueness and Overbreadth: Under the Fourteenth Amendment, the law was ruled unconstitutionally vague because it required speakers to guess at the “risk” of their content rather than following clear, objective standards.
• Compelled Speech: The requirement to include a state-mandated disclaimer was found to impermissibly alter the intended message and effect of the political speech.

Frequently Asked Questions

Why did the court rule against a law meant to stop deceptive AI content?

While the state has a valid interest in protecting elections, the court found that Act 191 was “substantially overbroad”. Instead of focusing solely on clearly harmful misinformation, it created a system in which creators of political satire or commentary had to assess vague risks of “harming reputation” before publishing. The court noted that less restrictive methods, such as electoral literacy campaigns or existing defamation laws, could address these concerns without infringing on free speech.

Does this ruling mean all AI-generated political content is now unregulated?

No, this specific ruling applies to the enforcement of Hawaii’s Act 191. The court suggested that the state could use speech-neutral alternatives to combat deepfakes, such as targeted counter-speech or enforcement of laws related to actual harm or defamation, rather than broad preventive restrictions.

What was the issue with the disclaimer requirement in Act 191?

The court determined that forcing a speaker to use a specific disclaimer is a form of compelled speech. In the context of political satire or parody, such a disclaimer would fundamentally change the content and the message the creator intended to convey, which is a violation of First Amendment protections.