AI: Minnesota Accuses Grok of Facilitating Digital Sexual Violence

Minnesota is taking direct action against Grok. The U.S. state accuses the AI from xAI of facilitating the creation of realistic sexual images from photos of identifiable individuals. A new law now prohibits such tools in the state. xAI wants to have it suspended. The conflict is now playing out in court.

In brief

• Since August 1, Minnesota has been enforcing a law against AI tools used to create fake sexual images.
• xAI is contesting the law on the grounds of the First Amendment.
• Grok has already been the subject of investigations and complaints regarding sexual deepfakes involving minors.

The AI Grok Directly Targeted by Minnesota

The new law targets platforms and software capable of generating realistic sexual images of an identifiable person from an existing photo. Grok finds itself on the front lines. In France, justice is already investigating X and Grok following reports concerning sexual deepfakes and images of children.

The law is numbered HF-1606. It was adopted in the spring with 132 votes in favor and one against in the Minnesota House. The Senate approved it unanimously, with 65 votes to zero.

Since August 1, the companies involved face fines of up to $500,000 per image in case of violation. Lawmakers heard the story of a man who used photos published on social media to create fake sexual images of over 80 women he knew. Minnesota now wants to cut off access to tools that allow such manipulation. xAI disagrees.

xAI Invokes Freedom of Speech

Elon Musk's company filed a lawsuit in July to prevent the law from being enforced. Its main argument is based on the First Amendment of the U.S. Constitution. xAI believes that the Minnesota law covers an excessively broad scope and could affect images that are otherwise protected by freedom of speech.

The company specifically cites photos of people shirtless, swimmers, or certain forms of political satire. It also criticizes the lack of exceptions for developers trying to limit abuse. Even an image created with the consent of the person concerned could fall under the law's provisions.

Minnesota Attorney General Keith Ellison defends exactly the opposite. For him, the law primarily regulates a tool that allows for the production of sexual images without consent. He accuses Grok Imagine of having significantly facilitated this type of practice.

The standoff already resembles those engaged with other authorities. In May, Elon Musk publicly criticized French magistrates investigating X and Grok. This time, xAI is directly trying to overturn the law.

Grok Already Faces Multiple Cases

Minnesota is not entering uncharted territory. In January, a watchdog organization estimated that Grok had generated over 23,000 sexualized images of children in just eleven days. Several countries then opened investigations. In March, three minors in California joined a lawsuit against xAI. They accuse Grok of having been used to transform their photos into AI-generated child exploitation content.

xAI claims to have reacted. The company states that it suspended over 50,000 accounts in 2026 and reported more than 70,000 cases to the National Center for Missing and Exploited Children.

Grok had already accumulated other controversies. In 2025, xAI acknowledged a mistake after the dissemination of hateful content by its AI. The Minnesota case goes further. It is not only asking xAI to correct Grok after an incident. It seeks to impose limits directly in the law. The courts will now have to decide whether Minnesota is regulating a dangerous tool or encroaching on freedom of speech protected by the Constitution. Grok remains available. The law does too.