xAI Sues Minnesota Over AI “Nudification” Law
xAI, a company founded by Elon Musk, has initiated a lawsuit against Minnesota regarding a groundbreaking state law that aims to ban the use of “nudity” techniques in websites and applications. This legal challenge could potentially set a precedent about how much states can constitutionally regulate artificial intelligence.
The lawsuit was filed in federal court on Monday, just a few days before the law is set to take effect on Saturday. If implemented, Minnesota would become the first state to try and prohibit the use of AI technologies designed to produce fake nude images of real individuals. The law was officially signed in May.
In its 38-page complaint, xAI acknowledged the state’s aim to prevent the unauthorized distribution of AI-generated nude images but argued that the law “overreaches.” It claims this legislation would ban numerous images and videos that fall under constitutional protection, imposing hefty fines—up to $500,000 for each violation.
The lawsuit points out the absence of a “safe harbor” provision for companies that actively attempt to stop users from creating such images. It also includes concerns that the law encompasses images that are either consented to by those depicted or created by them.
Additionally, xAI criticized the law for having an overly broad definition of “intimate parts,” asserting that it encompasses body parts that are frequently visible in public settings.
Minnesota’s Attorney General, Keith Ellison, stated that his office hasn’t yet seen or been officially served with the lawsuit. He expressed his concerns about the frightening implications of AI being used to generate nude images against individuals’ wishes. According to him, this misuse of AI strips individuals of their dignity, impacting them in profound personal and professional ways.
There have been mixed results in courts regarding laws aimed at regulating election-related deepfakes. A California law was blocked citing First Amendment rights. Interestingly, Minnesota’s law previously survived a court challenge from Musk’s company back in 2025.
This particular law sets itself apart from others that criminalize deepfakes, which are often defined as sexually explicit images created without consent. Most of those laws penalize the individuals who produce these images rather than the developers of the tools used to make them.
Moreover, a federal law enacted last year mandates that platforms have systems in place to eliminate such images. xAI argues that Grok, its AI chat and image generator, already has terms that prohibit activities like altering images to create nude appearances or produce sexually explicit content.
xAI emphasized that it enforces these rules by suspending accounts and reporting any suspected child sexual exploitation. Earlier this year, the company reiterated its commitment to not tolerating any form of non-consensual nudity or unwanted sexual content.
In response to global criticism concerning sexualized imagery, Grok announced in January that it would implement geo-blocking to restrict content based on local laws. It also committed to prohibiting the editing of images of real people to expose their clothing where such actions are illegal.
Interestingly, Grok has a feature called “Spicy Mode” that permits users to create explicit content, which has garnered disapproval from various governments around the world. In January, xAI claimed it had taken technical measures to prevent Grok accounts from editing images of people in revealing clothing.
