xAI, the artificial intelligence company owned by SpaceX, has filed a federal lawsuit challenging a Minnesota law that would ban so-called “nudify” applications, setting the stage for another high-profile legal battle over the limits of AI regulation, free speech protections and the growing threat of AI-generated sexual abuse.

The lawsuit, filed Monday in federal court in Minnesota, names Keith Ellison as the defendant and argues that the state’s new law violates constitutional protections by imposing sweeping restrictions on AI-powered image generation.

In its complaint, xAI said the statute “imposes an overbroad, content-based ban on free speech and the tools of visual expression in a clumsy attempt to prohibit ‘nudification.'”

The Minnesota law, scheduled to take effect on Saturday, targets applications and websites that enable users to create non-consensual sexually explicit images using artificial intelligence. Companies whose platforms are used to generate prohibited deepfakes could face civil penalties of $500,000 for each violation, making it one of the toughest state-level AI enforcement measures in the United States.

The legislation was introduced by Erin Maye Quade after learning about a case in which a man allegedly used publicly available social media photographs to generate sexually explicit AI images and videos of more than 80 women without their knowledge or consent.

Supporters of the law argue that it fills a growing gap in existing criminal and civil protections as generative AI dramatically lowers the barriers to creating convincing synthetic intimate images.

xAI contends, however, that the law reaches far beyond its intended purpose.

According to the company, the penalties are so severe that a platform whose users generated 100,000 prohibited images could theoretically face $50 billion in fines, an outcome the company described as unconstitutional and economically unsustainable.

The lawsuit centers on the First Amendment, with xAI arguing that the legislation unlawfully restricts protected forms of expression by regulating the technology itself rather than punishing unlawful conduct.

The case is expected to become an important test of how U.S. courts balance constitutional free speech protections against growing efforts by lawmakers to regulate harmful uses of generative AI.

Compounding Sexual Images Legal Battles

The legal challenge also comes as xAI confronts broader scrutiny over how its AI systems are used.

The company is already facing a proposed class-action lawsuit alleging that its Grok chatbot and image generation tools were used to create child sexual abuse material (CSAM) and AI-generated sexual images based on real individuals without their consent. The lawsuit further alleges that xAI failed to report information about users responsible for generating the images to law enforcement authorities.

xAI has denied wrongdoing and says its policies explicitly prohibit users from creating nude or sexually explicit images of individuals without their consent. The company said in its Minnesota complaint that it has implemented extensive technological safeguards designed to prevent such content from being generated and has pursued legal action against users who deliberately bypass those protections.

AI developers are increasingly being confronted by moral and legal challenges. As image-generation models become increasingly sophisticated, companies face growing pressure to prevent misuse while preserving legitimate creative and commercial applications of the technology.

Lawmakers across the United States have accelerated efforts to regulate AI-generated deepfakes following a surge in cases involving non-consensual sexual imagery, political misinformation, financial fraud and identity theft. Regulators now see synthetic media as one of the most immediate public safety risks associated with generative AI because the technology enables highly realistic content to be produced quickly and at minimal cost.

Supporters of Minnesota’s law say that traditional legal remedies have struggled to keep pace with the speed and scale at which AI-generated abuse can spread online. Maye Quade compared the legislation to longstanding laws prohibiting voyeurism and the creation of intimate images without consent, arguing that the statute modernizes existing protections for the AI era rather than creating entirely new restrictions on speech.

In an emailed statement, she defended the legislation, saying it protects First Amendment rights while addressing image-based sexual abuse.

“It’s sad that the creators of nudification technology would rather protect their ill-gotten profits than protect us from image-based sexual abuse,” she said.

Minnesota Governor Tim Walz, a Democrat, also responded publicly to the lawsuit, writing on social media: “See you in court, creep.”

The Minnesota case is not xAI’s first attempt to challenge AI-related regulation.

The company previously sued to block a California law aimed at restricting AI-generated deepfakes in elections. That challenge achieved a partial victory after a federal judge struck down provisions that would have limited certain forms of AI-generated election content on First Amendment grounds.

The latest lawsuit underscores the increasingly complex legal challenges facing AI developers. As states move more aggressively than Congress to regulate emerging AI technologies, courts are likely to play a central role in determining where constitutional protections end and government oversight begins.

Amazon-owned autonomous vehicle company Zoox has secured a major regulatory breakthrough after U.S. safety regulators granted it a temporary exemption that allows the company to begin charging passengers for rides in its purpose-built robotaxis, moving the self-driving startup significantly closer to large-scale commercial deployment.

The exemption, announced on Thursday by the National Highway Traffic Safety Administration (NHTSA), removes one of the final federal regulatory obstacles standing between Zoox and a fully commercial robotaxi business. The decision permits the company to operate its uniquely designed autonomous vehicles for paying customers despite the vehicles not complying with several long-standing federal motor vehicle safety standards that were written for conventionally driven cars.

Unlike traditional automobiles, Zoox’s robotaxis were designed from the ground up for autonomous driving and therefore do not include steering wheels, pedals, or a driver’s seat. To accommodate the unconventional design, NHTSA granted exemptions from eight federal safety standards, including regulations covering windshield defrosting systems and braking requirements for light-duty vehicles.

The exemption follows an earlier approval granted nearly a year ago that allowed Zoox to operate its robotaxis on public roads and provide free rides to passengers in cities including San Francisco and Las Vegas. That authorization, however, prohibited the company from charging fares, limiting the commercial viability of its operations.

The latest decision changes that equation by enabling Zoox to generate revenue from its autonomous ride-hailing service, representing a significant step toward validating Amazon’s multibillion-dollar investment in autonomous mobility.

The approval is subject to several conditions. Zoox may deploy up to 2,500 commercial robotaxis annually over the next two years and will operate under what NHTSA described as an “enhanced, adaptable oversight structure” designed to evolve alongside the company’s technology.

A Zoox spokesperson said the company plans to begin charging customers in Las Vegas in the near future, with additional cities to follow as state-level regulatory approvals are secured.

California, where Zoox is headquartered and already conducts extensive autonomous vehicle testing, remains an important target market. Before launching paid driverless rides there, the company must still obtain commercial deployment permits from both the California Department of Motor Vehicles and the California Public Utilities Commission.

Zoox Chief Executive Aicha Evans described the approval as a landmark moment for both the company and the broader autonomous vehicle industry.

“We are honored to receive the first-ever commercial exemption for a purpose-built robotaxi from NHTSA, enabling us to begin charging for our service and take another step toward bringing autonomous ride-hailing to more communities,” Evans said.

The exemption is notable because it marks the first time NHTSA has approved commercial deployment of a purpose-built robotaxi that was never intended to be driven by a human. Most autonomous vehicle developers, including Alphabet’s Waymo and Tesla, rely on modified versions of conventional passenger vehicles that retain traditional driving controls.

For Zoox, the decision validates a strategy that differs fundamentally from many competitors. Rather than adapting existing vehicles, the company has invested years developing a bidirectional, fully autonomous vehicle specifically optimized for ride-hailing, featuring symmetrical seating, no driver’s compartment, and the ability to travel equally well in either direction without turning around.

Alongside the Zoox announcement, NHTSA unveiled updates to its exemption framework that will make it easier for automakers to temporarily sell limited numbers of non-compliant vehicles while testing emerging technologies.

The agency also announced a partnership with SAE Industry Technologies Consortia to establish a three-year, $5 million initiative aimed at collecting safety data and accelerating the development of national autonomous vehicle performance standards.

The initiative seeks to create a unified national framework for evaluating autonomous driving systems, an area that has long been fragmented by differing state regulations and evolving federal oversight.

NHTSA Administrator Jonathan Morrison said the agency remains committed to supporting innovation without compromising public safety.

“By removing unnecessary barriers to innovation, developing industry guidance, and providing strong enforcement oversight while we create performance requirements, NHTSA is taking a balanced approach to AV regulation,” Morrison said. “These advancements will ensure that the United States continues to lead the world in AV technology in a safe and responsible manner.”

The announcements also point to the Trump administration’s supportive stance toward autonomous vehicle development as Washington seeks to strengthen U.S. leadership in next-generation transportation technologies.

NHTSA also disclosed that it is reviewing a separate exemption request from Los Angeles-based startup Robomart, whose autonomous delivery vehicle is designed to transport up to 500 pounds of goods. The agency said it will seek public comment on that application after completing its preliminary evaluation.