The war over deepfakes and campaigns is heating up, and the fakers have won an early victory with a federal judge’s ruling against California’s wide-ranging law banning manipulated materials about politicians in the months before an election.

U.S. District Judge John A. Mendez said the law was poorly drafted and the First Amendment protects a wide range of speech about government officials, even when they are “deliberate lies.”

While acknowledging the potential for abuse in deepfakes, Judge Mendez issued a preliminary injunction Wednesday against enforcing California’s law.

---

---

He called the law “a blunt tool that hinders humorous expression and unconstitutionally stifles the free and unfettered exchange of ideas.”

“While California has a valid interest in protecting the integrity and reliability of the electoral process, AB 2839 is unconstitutional because it lacks the narrow tailoring and least restrictive alternative that a content based law requires under strict scrutiny,” Judge Mendez said.

The ruling is a blow to Gov. Gavin Newsom, a Democrat who demanded the law after a satirical video used altered audio of Vice President Kamala Harris to make it seem like she was making inane and politically damaging comments.

The maker of that video, Christopher Kohls, who goes by MrReaganUSA and Mr. Reagan on social media, brought the case that Judge Mendez used to blaze new ground in First Amendment law.

Eugene Volokh, a leading First Amendment scholar, said the case is the first to test government officials’ efforts to police deepfake misinformation in politics.

“The judge did seem very skeptical about attempts to restrict even knowing falsehoods in election campaigns,” Mr. Volokh said. “I’m not sure that other courts will agree.”

Mr. Kohls’ video that inspired the deepfake law went viral when Elon Musk shared it on his X account, drawing 100 million views.

In the video, actual footage of Ms. Harris is combined with stitched-together audio that makes her sound like she is mocking her own campaign and denigrating President Biden.

At one point, the video shows her acknowledging “total incompetence” and calling herself “the ultimate diversity hire.” It also includes footage of actual Harris gaffes, such as boasting about a U.S. alliance with North Korea.

Mr. Newsom was so outraged that he took to social media to say the video should be illegal.

The Democratic-dominated California State Legislature quickly passed a bill, with more than two-thirds support in each chamber, making an emergency law to forbid “deceptive” content that harms a candidate’s “electoral prospects” within 120 days of an election.

The Legislature carved out an exception for parodies and satire but said those must be accompanied by large, explicit disclaimers throughout the video.

In passing the legislation, the Legislature said it was grappling with the “first-ever artificial intelligence (AI) election” and feared technology was spiraling out of control.

“In a few clicks, using current technology, bad actors now have the power to create a false image of a candidate accepting a bribe, or a fake video of an elections official ‘caught on tape’ saying that voting machines are not secure, or generate an artificial robocall in the governor’s voice telling millions of Californians their voting site has changed,” the Legislature said.

Another law, AB 2655, requires social media companies to police their platforms for anything that violates the new standard.

That law is being challenged in a lawsuit by The Babylon Bee, a website that parodies politicians, mainly liberals.

Mr. Volokh, a scholar at several prominent universities and founder of the influential legal forum Volokh Conspiracy, said Mr. Kohls’ video was so clearly a parody that he doubted it would have violated California’s law in the first place.

“It’s just not materially deceptive under the law because a reasonable viewer would recognize it’s not factual,” he said.

Judge Mendez seemed troubled by the vague definitions the law used to describe illegal activity.

“Almost any digitally altered content, when left up to an arbitrary individual on the internet, could be considered harmful,” he wrote.

He halted all enforcement, saying the statute was likely invalid.

He said the Supreme Court’s ruling in the 1964 Sullivan case established that the First Amendment protects political speech about the government — and, by extension, political campaigns — even when it includes “deliberate lies.”

“These same principles safeguarding the people’s right to criticize government and government officials apply even in the new technological age when media may be digitally altered,” he wrote. “Civil penalties for criticisms on the government like those sanctioned by AB 2839 have no place in our system of governance.”

Mr. Musk cheered the ruling on social media.

“Score one for the people’s right to free speech,” he said.

The Washington Times has sought comment from Mr. Newsom and California Attorney General Rob Bonta.

Mr. Volokh said the last time the Supreme Court touched on this area of law was in the 2012 Alvarez decision, which invalidated a federal “stolen valor” law that banned people from wrongly claiming they had earned military honors or decorations.

In that case, the justices ruled that merely being false doesn’t put a statement beyond the protections of the First Amendment. Mr. Volokh said it’s another step to rule lies are protected to the same extent as true speech.

He said some lower-level courts have struck down laws banning lies in campaign contexts, but the emerging use of artificial intelligence raises new questions.

“Maybe some courts will say it’s different when it’s a deepfake,” he said.