04 Oct 2024
A federal judge has blocked most of California’s new law preventing deepfake images from impacting elections, just two weeks after it was signed.
Law AB 2839 was signed by California Governor Gavin Newsom on 17 September. It was designed to prevent deceptive AI-generated content from impacting election outcomes
On Wednesday (02 October), Senior US District Judge John A. Mendez gave judgment in a case brought by Chris Kohls, suspending most of the law. Kohls argued that AB 2839 violated First Amendment rights where deepfakes are created as parody.
The judgment leaves only part of the law in force: the section requiring digitally-altered audio-only recordings to include a verbal disclosure that they were generated using AI
“Most of [the law] acts as a hammer instead of a scalpel,” the ruling states. It goes on to describe AB2839 as “a blunt tool that hinders humorous expression and unconstitutionally stifles the free and unfettered exchange of ideas” according to Politico.
Kohls, known on Twitter / X as ‘Mr Reagan,’ filed the suit to prevent enforcement of the new legislation after he posted an AI-generated video of a Kamala Harris campaign ad on X.
AB2839 was the toughest anti-deepfake law in the US to date; the judgment is clearly a setback for strong enforcement in the run-up to November’s presidential election.
But at least Elon Musk was pleased. He posted on X: “California’s unconstitutional law infringing on your freedom of speech has been blocked by the court. Yay!”
The battle to regulate deepfake imagery continues…