DOJ Supports xAI Challenge to Minnesota Nudification Law

Minnesota's HF 1606 took effect on August 1. The law prohibits owners or operators of websites, applications, software, and other services from allowing users to use AI to "nudify" or sexually alter images or videos, Gizmodo reported. Civil penalties can reach $500,000 for each unlawful access, download, or use.

According to the DOJ filing described by Raw Story, the federal government contends that HF 1606 defines protected anatomy more broadly than federal law's definition of an "intimate visual depiction." The filing cited the statute's treatment of areas including an inner thigh or breast, and argued that its coverage could extend to an artificially generated image of a shirtless man in a swimming pool where federal law would not apply.

The Justice Department's brief also warned that states should avoid measures that hinder US national and economic security, Raw Story reported. The filing followed a December 2025 executive order in which President Donald Trump directed the department to challenge state AI laws that, in the administration's view, burden American innovation.

## xAI's constitutional challenge

xAI's complaint argues that the statute is overly broad because liability does not turn on intent or the consent of the person depicted, according to Gizmodo. The Next Web reported that xAI describes HF 1606 as a strict-liability rule, under which platform liability depends on whether a user created a covered image rather than on a platform's knowledge or intent.

Gizmodo reported that xAI's policies prohibit users from generating nude or sexualized images of real people without consent. xAI's attorney Robert Dunn argued in court that the law lacks explicit exemptions for consent, artistic expression, or parody, according to The Next Web.

Judge Frank previously denied xAI's temporary restraining order request, citing the timing of the filing shortly before the law took effect, Gizmodo reported. That earlier ruling left the statute in force while the broader lawsuit proceeds.

Ellison's motion to dismiss points to reports that Grok generated nonconsensual sexual imagery after xAI introduced a photo-editing feature. Raw Story reported that the motion cited an estimate of 3 million sexualized images in 11 days, including roughly 23,000 involving children, during December 2025. Those figures are allegations cited in the state's court filing, not findings by the court.

The dispute places two policy questions in direct tension: protections against nonconsensual sexual imagery and the limits on state regulation of general-purpose AI systems. Companies operating comparable generative-media services face a growing compliance challenge when state definitions, liability standards, and federal rules differ. A ruling on the injunction could therefore be closely watched by developers whose image-generation products are available across multiple US jurisdictions.