Highlights

xAI filed a federal First Amendment suit against Minnesota Attorney General Keith Ellison to block HF 1606, the nation’s first law aimed at AI "nudification," which takes effect August 1. The statute makes platforms strictly liable for AI-generated realistic images that show body parts a person never exposed, with civil penalties up to $500,000 per image. xAI argues the law’s broad definition of “intimate part” and lack of safe harbors or scienter expose platforms to liability even when they attempt to prevent abuse or when images are consented to or never shared.

Sentiment Analysis

• The overall tone of the article is mixed and largely legalistic: it balances xAI’s free-speech challenge with Minnesota’s stated public-interest rationale. The piece presents factual details about the lawsuit, the statute’s provisions, and the events that prompted legislation, without overtly favoring either side. The emotional content centers on concern—both about potential harms from nonconsensual AI-generated sexual images and about sweeping restrictions that could chill legitimate expression. The legal arguments emphasize constitutional protection of tools and expression, while state officials emphasize victim protection and dignity.

Article Text

xAI, the artificial-intelligence company founded by Elon Musk, filed a federal lawsuit Monday seeking to block Minnesota’s HF 1606, the first U.S. law specifically aimed at AI "nudification." The statute, which is scheduled to take effect on August 1, creates strict liability for platforms when AI-generated images depict real people with exposed body parts those people never actually revealed. Each violation can carry civil penalties of up to $500,000 per image, a figure xAI highlights as punitive and potentially crippling for service providers and creators.

The complaint, filed in U.S. District Court for the District of Minnesota, frames the statute as a content-based restriction on speech and the tools of visual expression. xAI does not dispute Minnesota’s interest in preventing the dissemination of nonconsensual sexually explicit images; instead, the company argues the law reaches far beyond that goal. According to the suit, Minnesota adopted the term "intimate part" from a criminal sexual contact statute originally designed for physical, nonconsensual contact. That definition, xAI says, sweeps in disparate scenarios such as a shirtless politician, a person in a swimsuit, or a satirical image of a public figure—situations that implicate protected speech and legitimate uses of AI-generated imagery.

The complaint also focuses on the statute’s lack of safe harbors and scienter requirements. Under HF 1606, platforms would be liable even if they take active steps to prohibit nudification, filter content, or remove offending material. Liability would attach irrespective of whether the depicted individual consented, created the image themselves, or whether the image was ever distributed. xAI argues that this absence of a knowledge requirement and the failure to protect good-faith efforts to prevent harm mean ordinary companies could face severe penalties for isolated user misconduct.

In support of its position, xAI points to its recent enforcement efforts and moderation actions. The company says it suspended tens of thousands of accounts and reported thousands of incidents to law-enforcement entities in connection with problematic image generation. Those steps, the suit contends, would not insulate the company from liability under the Minnesota law. xAI’s legal complaints in other jurisdictions and its challenge to Colorado’s AI law earlier this year underscore the broader tension between new AI capabilities, regulatory responses, and constitutional protections.

The legislative backstory matters. Minnesota lawmakers passed HF 1606 overwhelmingly—132-1 in the House and 65-0 in the Senate—after a man used social-media photos to create sexual images of more than 80 women he knew. State advocates and lawmakers, including a bill sponsor who said limiting consensual images was intentional, emphasized the severe emotional and reputational harms inflicted by such nonconsensual imagery. Minnesota’s attorney general has defended the law’s purpose, arguing that AI nudification strips victims of dignity and causes deep personal and professional harm.

But xAI and its attorneys contend the state could have adopted a narrower framework. They point to the federal TAKE IT DOWN Act, signed in 2025, as an example: that law targets nonconsensual intimate deepfakes but requires proof of nonconsent and actual distribution, uses a narrower definition of covered body parts, and includes exemptions for educational or medical contexts. xAI argues Minnesota’s broader approach creates constitutional problems by imposing strict liability without distinguishing malicious actors from legitimate creators or consented activity. They warn the law would chill a wide range of protected expression and burden companies that provide general-purpose creative tools.

The dispute illustrates a wider policy dilemma: how to curb harms caused by emergent AI capabilities while preserving free expression and avoiding unduly burdensome liability for intermediaries. Courts will now consider whether Minnesota’s law is a permissible regulation to protect privacy and dignity or an unconstitutional restriction on speech and the tools used to make visual content. With HF 1606 set to take effect imminently, xAI is asking a federal judge to halt enforcement while the legal challenge proceeds.

Key Insights Table