Elon Musk’s xAI battles Minnesota over AI ‘nudification’ ban

The AI company said Minnesota's law is a content-based restriction that will penalize forms of protected speech, such as artistic expression or satirical parody.ST. PAUL, Minn. (CN) — In a federal courtroom Wednesday, xAI squared off with Minnesota over its first-in-the-nation legislation banning AI “nudification,” arguing state leaders overstepped their constitutional authority.In pursuit of a preliminary injunction against the ban, Elon Musk’s company told U.S. District Judge Donovan Frank state leaders certainly have a legitimate interest in fighting sexual harassment and AI-generated pornography, but Minnesota’s statute casts far too wide a net under the First Amendment.“I don’t know if the constitutional lawyers in the Legislature were out that day, but it’s clear the Legislature did not even consider the First Amendment problems,” xAI’s attorney, Robert Dunn, said. “The Legislature decided this problem was so severe it could pass constitutional scrutiny.”The statute challenged in xAI’s initial suit, HF 1606, took effect this month and prohibits anyone who owns or controls a website, app or software from allowing users to “nudify” or sexually alter images using AI generation.Violations of the statute carry penalties of up to $500,000.Frank previously denied xAI’s request for a temporary restraining order on July 31, allowing the statute to take effect. In his ruling, the Bill Clinton appointee cited the timing of xAI’s late filing — brought just days before the deadline — noting the delay undermined the claim that emergency relief was needed.Arguing for a broader injunction Wednesday, Dunn emphasized xAI already maintains strict platform policies prohibiting users from generating nude or sexualized images of real people without their consent — though the state said that’s not enough.“It’s certainly not surprising that xAI is not coming into federal court and saying, ‘We support these images,’” Minnesota Assistant Attorney General Janine Kimble said. “They say, ‘Look at our terms of service; we tell people not to do this.’ … And yet, they still are reporting tens of thousands of users creating images.”Kimble argued Grok — the generative AI chatbot developed by xAI — and other AI products have no barrier to entry, allowing users to run wild with inappropriate AI generation that is nearly impossible to prevent, remove or criminalize on a case-by-case basis.“The repercussions only happen after the fact,” Kimble said. “Someone has to see the image and report it; it needs to be findable; the user that created it needs to be a human being that is subject to the jurisdiction of a court, or able to be extradited if abroad. Lots of steps need to happen.”While reiterating its dedicated effort to remove users from its platform who violate the terms, Musk’s company said Minnesota’s law operates as an unconstitutional content-based restriction that risks penalizing swaths of protected speech.As the statutory text lacks explicit carve-outs for consent, artistic expression or satirical parody, xAI claims it could subject platforms to severe penalties for ordinary, lawful image modifications.“The state would punish xAI for allowing a user to nudify themselves,” Dunn said.Minnesota later countered, claiming studies show around 95% of all deepfake and manipulated images are nonconsensual and sexually explicit representations of others.“Without evidence from xAI about the scope of what they would call ‘non-bad nudification,’ you’re left with a record where the standard uses of this technology is harmful,” Kimble said.Musk’s company also claimed far less restrictive statutory language exists for Minnesota to achieve a nearly identical goal — most notably in the state’s deepfake and revenge porn policies, which address similar technological representation of “intimate parts.”“The Legislature has a very easy fix here,” Dunn said, arguing Minnesota should narrow its law to focus on the lack of consent and distribution of images while adding a “safe harbor” protection for platforms acting in good faith.“This thing passed unanimously; if they were to come back with a narrower statute, it would likely get passed unanimously too,” he said.While the state claims its ban is content-neutral and focused only on regulating the modification technology itself, xAI argues the regulation is based entirely on the image created, which treads dangerously close to censoring speech.“Our terms say we will kick you off the platform, and we do; we try to block it. … Some may sneak through,” he said. “But we’re being treated the exact same as any company that is deliberately doing this. Just telling us that we can’t allow certain types of speech to come out of our platform, that’s kind of a word game to pretend it’s a technology statute.”In a late Tuesday court filing, the federal government filed a motion as a third party in defense of Musk’s company, arguing that, while the dangers of AI nudification are certainly real and should be addressed, Minnesota’s law sweeps up constitutional conduct not prohibited under federal law.“As the nation remains in the earliest days of the AI technological revolution and in a race with adversaries for supremacy within it, states must be careful not to impose excessive measures that hinder American national and economic security," the government said in its motion, noting federal laws already exist addressing the same issues as Minnesota’s law.Frank took the motion for a preliminary injunction under advisement and said he would expedite a ruling as soon as possible.Subscribe to our free newslettersOur weekly newsletter Closing Arguments offers the latest about ongoing trials, major litigation and rulings in courthouses around the U.S. and the world, while the monthly Under the Lights dishes the legal dirt from Hollywood, sports, Big Tech and the arts.Additional Reads