- Регистрация
- 17 Февраль 2018
- Сообщения
- 26 053
- Лучшие ответы
- 0
- Баллы
- 2 093
Offline
One of California's new AI laws, which aims to prevent AI deepfakes related to elections from spreading online, has been blocked a month before the US presidential elections. As TechCrunch and Reason report, Judge John Mendez has issued a preliminary injunction, preventing the state's attorney general from enforcing AB 2839. California Governor Gavin Newsom signed it into law, along with other bills focusing on AI, back in mid-September. After doing so, he tweeted a screenshot of a story about X owner Elon Musk sharing an AI deepfake video of Vice President Kamala Harris without labeling it as fake. "I just signed a bill to make this illegal in the state of California," he wrote.
I just signed a bill to make this illegal in the state of California.
You can no longer knowingly distribute an ad or other election communications that contain materially deceptive content -- including deepfakes. https://t.co/VU4b8RBf6N
— Gavin Newsom (@GavinNewsom) September 17, 2024
AB 2839 holds anybody who distributes AI deepfakes accountable, if they feature political candidates and if they're posted within 120 days of an election in the state. Anybody who sees those deepfakes can file a civil action against the person who distributed it, and a judge can order the poster to take the manipulated media down if they don't want to face monetary penalties. After Newsom signed it into law, the video's original poster, X user Christopher Kohls, filed a lawsuit to block it, arguing that the video was satire and hence protected by the First Amendment.
Judge Mendez has agreed with Kohls, noting in his decision [PDF] that AB 2839 does not pass strict scrutiny and is not narrowly tailored. He also said that the law's disclosure requirements are unduly burdensome. "Almost any digitally altered content, when left up to an arbitrary individual on the internet, could be considered harmful," he wrote. The judge likened YouTube videos, Facebook posts and X tweets to newspaper advertisements and political cartoons and asserted that the First Amendment "protects an individual’s right to speak regardless of the new medium these critiques may take." Since this is merely a preliminary injunction, the law may be unblocked in the future, though that might not happen in time for this year's presidential elections.
This article originally appeared on Engadget at https://www.engadget.com/ai/judge-b...n-related-ai-deepfakes-133043341.html?src=rss
I just signed a bill to make this illegal in the state of California.
You can no longer knowingly distribute an ad or other election communications that contain materially deceptive content -- including deepfakes. https://t.co/VU4b8RBf6N
— Gavin Newsom (@GavinNewsom) September 17, 2024
AB 2839 holds anybody who distributes AI deepfakes accountable, if they feature political candidates and if they're posted within 120 days of an election in the state. Anybody who sees those deepfakes can file a civil action against the person who distributed it, and a judge can order the poster to take the manipulated media down if they don't want to face monetary penalties. After Newsom signed it into law, the video's original poster, X user Christopher Kohls, filed a lawsuit to block it, arguing that the video was satire and hence protected by the First Amendment.
Judge Mendez has agreed with Kohls, noting in his decision [PDF] that AB 2839 does not pass strict scrutiny and is not narrowly tailored. He also said that the law's disclosure requirements are unduly burdensome. "Almost any digitally altered content, when left up to an arbitrary individual on the internet, could be considered harmful," he wrote. The judge likened YouTube videos, Facebook posts and X tweets to newspaper advertisements and political cartoons and asserted that the First Amendment "protects an individual’s right to speak regardless of the new medium these critiques may take." Since this is merely a preliminary injunction, the law may be unblocked in the future, though that might not happen in time for this year's presidential elections.
This article originally appeared on Engadget at https://www.engadget.com/ai/judge-b...n-related-ai-deepfakes-133043341.html?src=rss
