Report from Straight Arrow News
In Brief – Federal Judge John Mendez has struck down California’s AB 2839 on summary judgement, ruling that the legislation combatting so-called political deepfakes violates the First Amendment. The measure, enacted in 2024, proposed to ban digitally manipulated communications that are false or misleading and target political candidates and election processes four months before an election. Mendez argued that the law unlawfully censors speech before any actual harm occurs, lacks precedent under the First Amendment, and could have been narrowly tailored to target only false speech causing real, legal harm. Last October, Mendez imposed a temporary injunction on that law to block it from being enforced during the 2024 election cycle. AB 2389 is the second California law intended to block misleading political content that Judge Mendez has blocked. In August, he imposed an injunction on AB 2655, which requires large social media companies to remove “materially deceptive content” about political candidates and election officials from their platforms. In announcing his ruling to block AB 2655, Mendez said from the bench that no part of that law was salvageable, and he took the opportunity to again express deep doubts about AB 2839 constitutionality, adding that the California State Legislature was told about clear First Amendment shortcomings with each measure and yet they went ahead and drafted them anyway.
Context – California passed a slew of AI-related bills in 2024. The highest profile was SB 1047, “AI safety” legislation often compared to the EU AI Act. Governor Newsome vetoed it after intense lobbying from the state’s venture capital and AI industries. However, the governor did sign AB 2655 and AB 2839, adding to the more than 20 US states that have enacted laws addressing election AI deepfakes. But none has been enforced to date. It is noteworthy that California’s election deepfake laws target misinformation and deception, not just instances using AI technology.