Minnesota starts enforcing a new law aimed at nudification apps
Minnesota began enforcing a new law on Saturday that targets so-called nudification tools, a category the statute defines as services that alter images of recognizable people to expose intimate body parts. The measure applies whether the tool uses artificial intelligence or another automated method, and it creates a notably aggressive liability structure for companies that provide the technology.
According to the supplied source text, the law allows victims to sue app providers and authorizes fines of up to $500,000 not only for generating an altered image, but for each time that image is accessed. That detail sharply raises the financial stakes for platforms whose outputs can spread quickly online. A single manipulated image that is repeatedly viewed or shared could create exposure far beyond a one-time penalty.
The law took effect after U.S. District Court Judge Donovan Frank denied a late request from xAI for a temporary restraining order that would have blocked enforcement. The source text says xAI, described there as the originator of the Grok AI chatbot, argued that the law is broad enough to capture some image generations that are not clearly malicious. But the judge’s reasoning, at least in part, turned on timing. He noted that xAI filed its motion on July 29, 2026, nearly three months after the law was signed and only three days before it was set to take effect, writing that such a delay suggested the claimed harm was not immediate.
How the statute defines nudification
The supplied text says Minnesota defines nudification as altering a photo of a recognizable person to expose “intimate parts.” It further states that the term “intimate parts” draws from a separate Minnesota statute covering the primary genital area, groin, inner thigh, buttocks, or breast of a human being. The law is also described as focusing on image alteration that can be done essentially at the push of a button, rather than work that involves substantial manual artistry or skill.
That distinction matters because it frames the measure as an attempt to target scalable, consumer-facing manipulation tools rather than all forms of digital editing. In practice, the policy appears tailored to the recent wave of systems that can produce sexualized fake imagery with little user effort and at large volume. The narrower focus on low-friction generation may also reflect an effort to connect the law to a specific, emerging product category instead of a broader challenge to image editing itself.
Even so, the source text indicates that xAI’s lawsuit attacks the law as overbroad. The company argues that the definition could sweep in synthetic images made in contexts it characterizes as innocent or satirical. The examples cited in the suit, according to the source text, include AI-generated images of Trump administration officials in bathing suits and another image depicting Chris Christie and J.D. Pritzker in what were described as sumo mawashi.
A response to rapidly scaled image abuse
The larger context is the speed with which generative systems have made non-consensual sexualized imagery easier to create and distribute. The supplied text says the Grok AI chatbot became associated during the most recent holiday season with the non-consensual generation of sexualized or scantily clad images of people. It also cites an estimate that Grok created 23,000 such images of children. That figure, if accurate, helps explain why lawmakers are moving toward product-level liability rather than relying only on existing revenge porn or harassment frameworks.
Minnesota’s approach is notable because it appears to put responsibility on the provider of the tool, not just the end user. That matters in the current AI market, where platforms often emphasize scale, speed, and easy access. A law that can impose steep penalties for every access of a manipulated image changes the incentives. Providers may need stronger safeguards before release, faster response systems after abuse is detected, and tighter restrictions on prompts, uploads, or public sharing features.
The source text says X posted a statement indicating that the capabilities enabling this behavior had been curtailed. That suggests at least some platform operators already see moderation and product controls as unavoidable. But Minnesota’s law pushes that logic further by turning safety failures into a potentially direct legal and financial problem.
Why this case could matter beyond Minnesota
The immediate effect is local, but the implications are broader. Legislatures elsewhere are grappling with how to regulate generative tools that can produce realistic abusive material at scale. Minnesota’s model appears to combine a concrete product definition with private enforcement and large civil penalties. If it survives future legal challenges, it could become a template for other states looking for a faster route than broad federal action.
The legal dispute also highlights a familiar tension in AI governance. Companies often argue that broadly written rules could catch legitimate expression, parody, or experimentation. Lawmakers and victims’ advocates argue that the technology’s social cost is no longer hypothetical, especially when systems can create harmful material involving ordinary people with almost no technical skill required. Minnesota is, at minimum, signaling that automated nudification belongs in a higher-risk regulatory bucket.
What comes next will likely turn on both litigation and compliance. xAI failed to stop the law before it took effect, but that does not mean the broader constitutional or statutory fight is over. For now, though, Minnesota has moved from debate to enforcement. In a policy area where many governments are still studying the problem, that alone is significant.
This article is based on reporting by Gizmodo. Read the original article.
Originally published on gizmodo.com
