X.AI LLC v. Ellison (challenging Minnesota's nudification-technology statute)

X.AI LLC v. Ellison (Civil No. 26-3425 (DWF/DTS), D. Minn., filed July 27, 2026) is a First Amendment challenge brought by xAI against Minnesota Attorney General Keith Ellison, in his official capacity, to Minn. Stat. § 325E.91 — the statute enacted as H.F. 1606, which prohibits an operator of a website, application, software, program or other service from allowing a user to nudify an image or video, or nudifying one on a user's behalf. The challenge is pleaded both facially and as applied, and the product at issue is Grok Imagine, the image and video generation model in the Grok family (Memorandum Opinion and Order Denying Preliminary Injunction — X.AI LLC v. Ellison (D. Minn., September 4, 2026)).

The court has twice declined to enjoin the statute — a temporary restraining order on July 31, 2026 and a preliminary injunction on September 4, 2026 — in each case on the movant's delay and failure to show irreparable harm rather than on the strength of the constitutional claim, which remains undecided.

The statute at issue

Section 325E.91 sits in the Minnesota commerce chapter rather than the criminal code, and attaches liability to the operator of a tool rather than to the person who creates or publishes an image. Subdivision 2 bars an owner or controller of a service from allowing a user to access, download or use it to nudify an image or video, and from nudifying on a user's behalf. Subdivision 3 exempts services that "require the technical skill of a user" to nudify. Subdivision 1(d) defines nudification as a process by which an image or video "is altered or generated to depict an intimate part not depicted in an original unaltered image or video of an identifiable individual" and where the result "is so realistic that a reasonable person would believe that the intimate part belongs to the identifiable individual"; "intimate part" takes the meaning given in Minn. Stat. § 609.341, subd. 5. The attorney general may enforce the section and seek civil penalties of up to $500,000 for each violation, together with injunctive and other relief, under § 325E.91, subd. 5 and Minn. Stat. § 8.31, subd. 3. The bill text is summarized at Minnesota HF 1606 — Prohibition on Nudification Technology (bill text, 2nd engrossment).

The opinion records the enactment sequence by citation to the legislative journals: the Minnesota House passed the bill on April 23, 2026 (Minn. H.J., 94th Leg., Reg. Sess. 6139-40 (2026)), the Senate passed it unanimously on April 29, 2026 (Minn. S.J. 9257-58), the Governor signed it on May 7, 2026 (Minn. H.J. 7315), and it took effect August 1, 2026 (Memorandum Opinion and Order Denying Preliminary Injunction — X.AI LLC v. Ellison (D. Minn., September 4, 2026)). Contemporaneous coverage dates the House vote to April 28, 2026 and reports it as 132–1; the discrepancy between the journal citation and the reported date is unresolved and is recorded rather than reconciled at Minnesota HF 1606 (Nudification Technology Prohibition).

Procedural history

xAI filed the complaint on July 27, 2026, nearly three months after the bill was signed and four days before the statute took effect, and moved for a temporary restraining order. On July 31, 2026 the court denied that motion because of the delay in bringing it, indicating instead that it would hear the request as one for a preliminary injunction (Doc. No. 21). The State opposed (Doc. Nos. 20, 38). On September 4, 2026 Judge Frank denied the preliminary injunction in a fourteen-page memorandum opinion and order (Doc. No. 54).

On August 17, 2026 the United States Department of Justice filed a statement of interest largely adopting xAI's position, writing that the statute "sweeps up constitutional and productive conduct" and gives platforms no opportunity to remove images before penalties attach, while stating that the department supports Minnesota's interest in prohibiting AI-generated sexual abuse material (Source: politico.com). The September 4 opinion does not address the filing.

xAI stated in a court filing the same day that it will appeal to the Eighth Circuit (Source: reuters.com).

The parties' positions

xAI argues that the statute "imposes an overbroad, content-based ban on free speech and the tools of visual expression," while stating that it "does not contest Minnesota's interest in prohibiting the dissemination of artificially generated nude images of real people without their consent" and contending that the law "extends far beyond that goal" (Source: cbsnews.com). It has also argued that "although Minnesota likely has a compelling interest in preventing individuals from creating nudes of other people without consent and distributing that content, the statute punishes AI platforms for assisting in the creation of a substantial amount of protected speech in relation to that limited goal" (Source: mprnews.org). The complaint pleads that Grok Imagine outputs span "art, animated family photographs, memes, product mockups, political satire, and religious imagery," that its acceptable-use policy prohibits nudifying a real person's image and sexualizing children, and that in 2026 alone xAI suspended over 50,000 accounts and made over 70,000 reports to the National Center for Missing & Exploited Children, resulting in at least 244 arrests.

The State challenges xAI's standing to assert a First Amendment claim on behalf of its users, disputes that xAI has a First Amendment interest of its own at stake, and argues that the statute is valid under any level of scrutiny. It also argues that xAI's existing measures are insufficient, citing three suits alleging that photographs of children were used to create child pornography on Grok Imagine — Doe 1 v. X.AI Corp., No. 5:26-cv-2246 (N.D. Cal., amended complaint July 7, 2026), and Does v. X.AI, Nos. 4:26-cv-750 and 4:26-cv-772 (E.D. Ark., amended complaints August 17, 2026), treated together at Grok CSAM suits (Doe plaintiffs v. X / xAI) — and noting that SpaceX, which the opinion describes as xAI's parent company, has set aside more than $500 million for potential litigation losses stemming from Grok Imagine's "Spicy" and "Unhinged" modes (Memorandum Opinion and Order Denying Preliminary Injunction — X.AI LLC v. Ellison (D. Minn., September 4, 2026)).

The preliminary-injunction ruling

The court denied relief on two independent grounds and expressly declined to reach the merits.

Delay and irreparable harm. The opinion notes that the statute was signed on May 7, 2026 and that xAI moved "nearly three months later and only three days before the law was set to take effect," offering no explanation beyond the suggestion that it takes time for such laws to reach corporate management. The court found there was "no serious dispute that xAI is a sophisticated and well-resourced litigant with the means to move quickly when its interests are at stake," and that the failure to act sooner "strongly suggests a lack of irreparable harm."

On the harm itself, the court treated the $500,000-per-violation penalty as monetary and therefore recoupable if the statute were later invalidated, and observed that penalties were unlikely in any event because xAI had already disabled its nudification tool in Minnesota. It found the claimed commercial injury unsupported: the geographically targeted controls xAI built into Grok Imagine's image-editing pipeline between July 24 and August 1, 2026 were a cost already incurred with no evidence of future cost; the assertion of substantial diverted engineering resources was "vague and conclusory"; and the projected loss of Minnesota users from a moderation rate risen "over 10 percent" was "purely speculative," resting on a single declaration by product counsel team lead Naser Baseer. The court added that because the statute applies to every company providing nudification-capable tools, "there will be no loss of users vis-à-vis any other competing company."

xAI's argument that a First Amendment violation alone establishes irreparable harm drew the fullest treatment. Citing the Eighth Circuit's decision in Maixner v. Eidsness, 182 F.4th 736 (8th Cir. 2026), the court noted that the rule applies where First Amendment interests are threatened or being impaired at the time relief is sought, and that even then a movant must make a clear showing of likely irreparable harm — quoting the Supreme Court in Benisek v. Lamone for the proposition that in a First Amendment case "a preliminary injunction does not follow as a matter of course from a plaintiff's showing of a likelihood of success on the merits."

Balance of harms and public interest. Merging the two factors as required where an injunction is sought against the government, the court found the balance "tips steeply in favor of the State," which had "enacted, democratically and nearly unanimously, a civil statute aimed at protecting the public from the undisputed harm stemming from AI-nudification technology." It characterized the choice to regulate the provision of tools rather than user conduct as "a direct response to the unique harms stemming from the new technology and the State's conclusion that curtailing user violations after harmful images are circulated is not enough."

Merits reserved. The opinion states that "the constitutional issues raised by the parties are complex, particularly when considered in the context of this new technology and the risks that it poses to the public," that those issues "deserve, and will receive, full consideration," and that a determination of the constitutional claims "will be made in the near future because the State has filed a motion to dismiss that places these issues front and center."

Legislative record relied on

The opinion devotes several pages to the evidence before the Minnesota legislature, material that shapes the public-interest analysis. At a February 19, 2026 Commerce Finance and Policy Committee hearing, three women testified to harms following a perpetrator's use of nudification technology on their likenesses; that perpetrator created non-consensual pornographic material of more than 80 women. RAINN told the committee that in one month of 2023 there were 24 million users of nudification apps, and that in 2023 over 95 percent of manipulated images were nonconsensual, sexually explicit and depicted women. The Internet Watch Foundation reported 6,800 reports flagging generative-AI child sexual abuse material in the first half of 2024, rising to over 400,000 tips through June 2025, and identified a guide encouraging use of nudification tools to generate material and blackmail children. At a March 26, 2026 Judiciary Finance and Civil Law Committee hearing, RAINN submitted a study of the first 11 days after a social media platform launched a one-click AI image-editing tool, finding roughly 65 percent of images created were photorealistic sexualized images — about 3 million, at 190 per minute — of which roughly 23,000 depicted children (Memorandum Opinion and Order Denying Preliminary Injunction — X.AI LLC v. Ellison (D. Minn., September 4, 2026)).

Relation to other challenges

The case is one of several First Amendment challenges by xAI to state AI statutes, alongside xAI LLC v. Weiser (challenging the Colorado AI Act) against the Colorado AI Act and SpaceXAI v. Bonta (AB 2013 challenge). The Justice Department's participation on the challenger's side follows the pattern it set in the Colorado case and is discussed with the preemption material at State-Level AI Regulation. The overbreadth and content-neutrality questions the court reserved are surveyed at Algorithmic Speech Doctrine, and the design-code litigation at NetChoice v. Bonta (CAADCA litigation) raises adjacent questions about regulating a service's architecture rather than its speech. The underlying subject matter is treated at Synthetic Media / Deepfakes and State Deepfake Statutes (MN, WA, TX, CA), and the child-safety dimension at AI and Children.

Open questions

• Whether the court reaches the merits through the State's pending motion to dismiss or through a later permanent-injunction posture.
• Whether the Eighth Circuit appeal is taken from the preliminary-injunction denial and, if so, whether it addresses only the equitable factors or the constitutional claim.
• How the subdivision 3 technical-skill exemption is construed, since it determines whether general-purpose creative tools fall inside or outside the prohibition.
• Whether xAI's Minnesota-specific disablement of the nudification tool persists and how it bears on any as-applied claim.

Relationships

• instance-of: Algorithmic Speech Doctrine
• depends-on: Memorandum Opinion and Order Denying Preliminary Injunction — X.AI LLC v. Ellison (D. Minn., September 4, 2026)
• depends-on: Minnesota HF 1606 (Nudification Technology Prohibition)
• depends-on: Minnesota HF 1606 — Prohibition on Nudification Technology (bill text, 2nd engrossment)
• litigates: xAI
• related: xAI LLC v. Weiser (challenging the Colorado AI Act)
• related: SpaceXAI v. Bonta (AB 2013 challenge)
• related: NetChoice v. Bonta (CAADCA litigation)
• related: Grok CSAM suits (Doe plaintiffs v. X / xAI)
• related: State-Level AI Regulation
• related: Synthetic Media / Deepfakes
• related: AI and Children
• related: Grok (xAI)

Sources

1. Memorandum Opinion and Order, X.AI LLC v. Ellison, No. 0:26-cv-03425-DWF-DTS (D. Minn. Sept. 4, 2026), Doc. No. 54 — fingfx.thomsonreuters.com
2. Reuters, "Musk's xAI loses court bid to block Minnesota's AI 'nudification' ban" (2026-09-04) — reuters.com
3. CBS News Minnesota, "Elon Musk's xAI sues Minnesota over law banning AI nudification" — cbsnews.com
4. MPR News, "Elon Musk's company xAI sues Minnesota over new law" (2026-07-28) — mprnews.org
5. Politico, Digital Future Daily (2026-08-19) — politico.com