A federal appeals court has paused enforcement of Minnesota’s AI “nudification” ban, siding for now with Elon Musk’s xAI in a First Amendment fight that could decide whether creators can use state law to go after apps that generate fake nude images of them.
The Eighth Circuit Court of Appeals granted xAI’s request for a stay on 5 October, pending a full appeal, in a brief order that gave no explanation of its reasoning. The practical effect: Minnesota’s law, which carries civil penalties of up to $500,000 per violation, is not currently being enforced against xAI while the case plays out.
Why it matters to creators
If you’ve ever worried about your likeness being fed into an AI “nudify” tool, this case is the one to watch. Minnesota’s law was one of the sturdiest state-level weapons against exactly that: a civil right of action against any platform or app that lets users generate sexualised images of real people without consent. With enforcement on hold, that remedy is effectively off the table in Minnesota for as long as the stay lasts — and the First Amendment argument xAI is making could, if it succeeds, weaken similar laws other states are drafting or have already passed.
What the law actually did
Minnesota’s HF 1606 was signed by Governor Tim Walz on 7 May 2026 after passing the state House 132-1 and the Senate 65-0 — about as close to unanimous as state legislatures get. It took effect on 1 August 2026 and bars anyone who owns or controls a website, app or piece of software from letting users “nudify” or otherwise sexually alter images of real people through AI. Violators face civil liability of up to $500,000, on top of damages owed to the person depicted.
xAI — maker of the Grok chatbot and image generator — sued to block the law on 27 July, arguing it is an unconstitutional, content-based restriction on speech. “The statute casts far too wide a net under the First Amendment,” the company argued, saying the law lacks carve-outs for consenting adults, satire or artistic expression and that xAI already bans this kind of generation through its own terms of service.
US District Judge Donovan Frank wasn’t persuaded. He rejected xAI’s bid for a temporary restraining order in late July, and after Attorney General Keith Ellison’s office filed its opposition — arguing xAI had shown no irreparable harm, had sat on the lawsuit for weeks after the law passed, and was unlikely to win on the merits — the court again declined to block enforcement in early September. The US Department of Justice weighed in on xAI’s side during that round, arguing the state law could hamper national security and US competitiveness in AI, a federal intervention NudeNewz covered at the time.
What changed on 5 October
xAI took the fight to the Eighth Circuit, and this time it won — at least temporarily. The appeals court’s stay pauses enforcement while the full appeal is heard, which could take months. Ellison’s office isn’t backing down: a spokesperson reiterated that “Minnesota’s nudification ban outlaws AI technology products from generating sexual images that harm and harass people in the vilest way possible,” and the state says it will “vigorously defend the law” as the appeal proceeds.
Nobody should read this as the law being struck down — it’s a procedural pause, not a ruling on the merits, and the one-line order doesn’t tell us how the Eighth Circuit is leaning. But it’s a live reminder that the legal protections creators are increasingly told to rely on can be put on hold by a single appellate order, sometimes with zero warning.
The bigger pattern
This sits alongside a rougher few weeks for deepfake legislation generally. Senator Ted Cruz single-handedly blocked the federal NO FAKES Act from reaching a vote on 3 October, meaning there’s still no nationwide civil remedy for AI deepfakes of performers. That leaves creators dependent on a patchwork of state laws — exactly the kind of law now under pressure in Minnesota. UK creators have it slightly differently: the nudification and non-consensual “deepfake” offences that came into force under the Crime and Policing Act on 29 June are criminal offences prosecuted by the state, not a civil claim a tech company can challenge on First Amendment grounds — there’s no First Amendment equivalent in English law to litigate around, though the UK route depends on police and the CPS actually pursuing cases.
What to do now
- Don’t treat any single law — state, federal or otherwise — as your only line of defence against deepfakes. Layer protections: platform reporting tools, DMCA takedowns, and services like the ones covered in our deepfake scam guide all still work regardless of this case.
- If you find AI-generated sexual content of yourself, preserve it before it’s removed — screenshots, URLs and timestamps — in case you need evidence for a platform report or a future legal claim.
- US creators in states with similar nudify laws (several have followed Minnesota’s lead this year) should keep an eye on this appeal; an Eighth Circuit ruling against Minnesota could invite copycat challenges elsewhere.
- Report any AI chatbot or app impersonating you directly to the platform’s trust and safety team — most major platforms, including X and Meta, now have dedicated deepfake-reporting flows separate from general copyright complaints.
- UK-based creators should remember the Crime and Policing Act route sits outside this US litigation entirely and isn’t affected by how the Eighth Circuit rules.
This is general information, not legal advice — speak to a qualified lawyer about your specific situation, especially if you’re considering a claim under a state nudification law.
