Elon Musk Is Suing Minnesota for the Right to Keep Posting Deepfake Nudes, Child Porn, and Revenge Porn On X.

Yes, you read that correctly.

Let me make sure I have this straight, because every time I read the sentence my brain tries to reject it like a bad organ transplant.

Elon Musk’s AI company, xAI — the one that makes Grok — filed a federal lawsuit on Monday against the state of Minnesota. Not because Minnesota banned AI. Not because Minnesota taxed AI. Because Minnesota passed a law banning nudification technology — software that takes a photo of a real, clothed, non-consenting person and generates a fake nude of them. The law takes effect Saturday. xAI would like a judge to stop that from happening.

That’s it. That’s the lawsuit. The richest man on earth is spending lawyer money to fight for the legal viability of the “deep fake nude porn” button.

The law he’s suing over passed 132–1

In an era where our politics can’t agree that water is wet, Minnesota’s nudification ban passed the state House 132 to 1 and the state Senate 65 to 0. It was the first law of its kind in the country, written largely because nudification apps have become a school cyberbullying weapon — teenage girls finding out that fake explicit images of them are circulating among their classmates.

This is about as close to a moral consensus as American lawmaking produces anymore. Democrats, Republicans, everybody looked at “AI software that strips women and kids without consent” and said: no. One guy in the entire Minnesota legislature voted against it.

And xAI looked at that near-unanimous, protect-the-kids statute and said: see you in court.

What xAI is actually arguing (and why it’s still damning)

To be fair to the lawyers, because someone has to be: xAI is not literally arguing “we have a constitutional right to make fake nudes.” The complaint says the company doesn’t dispute that Minnesota can outlaw distributing nonconsensual AI nudes — it argues the law goes way beyond that. Their claims, roughly:

• It’s a content-based speech restriction, so it’s presumptively unconstitutional under the First Amendment unless narrowly tailored.

• The definition of “intimate parts” is too broad — they complain it covers things like inner thighs and portions of breasts, so a bathing-suit image could theoretically trip it.

• It’s strict liability with no safe harbor: xAI says it gets punished even though its own terms of service already prohibit nudifying real people, no matter what safeguards it builds.

Some of that is the kind of overbreadth argument lawyers make about every internet law, and courts do sometimes agree. Fine. Here’s the thing though: you don’t have to file the lawsuit. xAI chose to be the plaintiff. Of every AI company in America — and every one of them is theoretically affected by this statute — exactly one marched into federal court to kill it before it took effect. The one whose product just did the thing the law exists to prevent, at industrial scale.

Which brings us to the holidays.

The company suing for “free speech” just had a mass nudification event

Over this past winter break, a Grok image capability went viral in the worst way imaginable. Users generated a flood of nonconsensual sexual images of real people — and, per reporting, a huge number of the targets were minors. One estimate extrapolated that Grok had produced something like 23,000 sexualized images of children before xAI locked it down and put out the standard-issue “we have zero tolerance” statement.

Zero tolerance. The tolerance was, empirically, about 23,000.

So the timeline is: Grok becomes a global scandal for spitting out fake porn of real women and kids → Minnesota (and other states, and other countries) moves to ban exactly that technology → and xAI’s response is not “thank you, we clearly cannot be trusted to self-regulate,” it’s a federal lawsuit to strike the ban down, with a side offer to switch off Grok’s image editing for Minnesotans in the meantime. Minnesota AG Keith Ellison’s response was about as restrained as I could manage: there are plenty of worthy AI debates to have, and this is not one of them, because nudification robs people of their dignity and causes real harm.

Meanwhile, in France

Now zoom out, because the Minnesota suit is not happening in a vacuum. It’s happening while French prosecutors have escalated their probe of Musk and X into a full criminal investigation — and the alleged offenses are not parking tickets. Paris prosecutors are pursuing charges that include complicity in possessing and distributing child sexual abuse images, spreading nonconsensual sexually explicit deepfakes, unlawful data collection, and — because Grok also apparently generated Holocaust-denial content, which is a crime in France — denial of crimes against humanity. French police raided X’s Paris offices in February. Musk was summoned, skipped the voluntary interview, and got a formal summons instead. (These are allegations; nobody’s been convicted of anything. But “criminal investigation into complicity in distributing CSAM” is not a phrase most CEOs ever have attached to their name, allegedly or otherwise.)

Other jurisdictions are circling too, including the California attorney general. The through-line in all of them is the same question: did Musk’s companies knowingly let a deepfake porn machine run because it drove engagement and hype?

And here’s where it gets genuinely dark

Ask yourself what Musk has been doing politically while all of this piles up. He’s spent the last year and a half openly campaigning to blow up the governments trying to regulate him. He endorsed the AfD in Germany — “Only the AfD can save Germany” — and wrote them a love letter in a national newspaper. He’s attacked Keir Starmer relentlessly. He’s smeared European regulators as tyrants. And in December, days after the EU hit X with a €120 million fine — the first ever under the Digital Services Act — he declared that the European Union should be abolished entirely.

Now, in fairness to the timeline: Musk’s far-right European crusade started before the Grok deepfake scandal blew up. The AfD endorsement, the Starmer feud, the MEGA posting — that’s all early 2025. So I can’t tell you his motive is purely “install governments that won’t prosecute me.” Maybe it started as ideology and vibes.

But look at what it’s become. The pattern now is unmistakable: every institution that moves to hold him accountable — the EU with its fines, France with its criminal probe, Minnesota with its 132-to-1 law — immediately becomes, in Musk’s telling, a tyranny that must be dismantled, sued, or replaced with friendlier politics. The U.S. Justice Department even declined to cooperate with the French investigation, calling it politically motivated. When the accountability comes from abroad, delegitimize it. When it comes from a U.S. state, litigate it to death. When it comes from voters, fund the parties that promise to make it go away.

Whatever the mix of motives — ego, ideology, engagement metrics, legal self-preservation — the output is the same: the world’s richest man, whose chatbot allegedly generated tens of thousands of sexualized images of children six months ago, is now in federal court arguing that a near-unanimous law against undressing people without their consent violates his company’s speech rights, while simultaneously agitating for the collapse of the foreign governments investigating him for it.

There are plenty of worthy debates to have about AI. Ellison’s right. This is not one of them. And the fact that Elon Musk thinks it is — that this is the hill, of all hills — tells you everything about what that company is actually for.

Sources: MPR News, Minnesota Reformer, Bloomberg Law, Gizmodo, AP/France 24, CNBC, Euronews, Fortune.