Elon Musk’s AI company xAI filed a federal lawsuit on Monday against Minnesota Attorney General Keith Ellison, asking a judge to block what it describes as the first U.S. law aimed at AI "nudification" tools before it takes effect. In the complaint, xAI calls the measure "an overbroad, content-based ban on free speech and the tools of visual expression in a clumsy attempt to prohibit nudification."
The law at issue is HF 1606. It makes platforms strictly liable if users create realistic images of real people showing body parts those people never exposed. Under the statute, each violation can bring a penalty of up to $500,000 per generated image, regardless of whether the platform knew about the image or intended the result.
The challenge focuses on scope, not a defense of nudification itself
The complaint was filed in the U.S. District Court for the District of Minnesota. xAI says it is not challenging the state’s interest in stopping nonconsensual fake nude imagery as such. The filing states: "xAI accordingly does not contest Minnesota's interest in prohibiting the dissemination of artificially generated nude images of real people without their consent."
Its argument is that the statute goes well past that objective. According to the complaint, the law exposes a broad range of protected expression to civil liability and government sanctions. xAI says Minnesota imported the term "intimate part" from a criminal sexual contact law written for nonconsensual touching rather than image generation. Under that definition, the inner thigh, buttocks, and breast are treated the same.
xAI argues that this opens the door to liability for realistic AI images of a shirtless politician, a woman in a swimsuit, or a public figure at the beach, including material created for political satire.
No safe harbor and no knowledge requirement, xAI says
The company also argues that HF 1606 gives platforms no safe harbor for good-faith efforts to reduce abuse. The complaint says: "There is no safe harbor for good-faith efforts of the provider of general-purpose AI creative tools to avoid harms." In xAI’s telling, liability can still attach even if the person shown in the image consented, created the image themselves, or the image was never shared at all.
The lawsuit also says the statute includes no scienter requirement, meaning knowledge of wrongdoing is not needed for liability. A user who slips past a company’s own filters could still leave the platform fully exposed under the law.
To support that point, xAI says that in 2026 alone it suspended more than 50,000 accounts and filed more than 70,000 reports to the National Center for Missing & Exploited Children, or NCMEC, leading to at least 244 arrests. None of those actions would provide a defense under HF 1606, according to the complaint.
Grok’s image tools have already drawn regulators and lawsuits
The filing comes against a wider backdrop of scrutiny around xAI’s image products. Grok’s image tools ran into trouble soon after their late July 2025 launch, when they were found capable of producing deepfakes of real people from simple text prompts. That prompted regulatory probes across Europe, Australia, and multiple U.S. states.
xAI has also faced separate lawsuits over those outputs. Baltimore and three minors in Tennessee each sued the company over deepfake material generated by Grok. In April, xAI also challenged Colorado’s AI law on similar First Amendment grounds.
Minnesota passed the bill by wide margins
Minnesota’s legislature passed HF 1606 by a vote of 132-1 in the House and 65-0 in the Senate. The measure followed a case in which a man used social media photos to generate sexual images of more than 80 women he knew.
Senator Erin Maye Quade, who championed the bill in the Senate, confirmed that the law’s application to consensual images was "intentional."
xAI points to the federal TAKE IT DOWN Act as a narrower model. That law, signed by President Donald Trump in May 2025, already covers nonconsensual intimate deepfakes, but requires proof of nonconsent and actual distribution, uses a narrower definition of body parts, and includes carve-outs for educational and medical content. xAI says Minnesota should have followed that approach instead.
Ellison says the law addresses serious harm
Ellison has shown no sign of retreating. After the lawsuit was filed, he said: "AI nudification robs the target of their dignity and can cause immense harm on an emotional, personal, and professional level."
HF 1606 is set to take effect on Saturday. xAI is asking a federal judge to stop it before then.