Three Tennessee teenagers have filed a class action lawsuit against Elon Musk and his artificial intelligence company xAI, alleging that its chatbot Grok was used to generate millions of sexualized deepfake images — including approximately 23,000 depicting children — over an 11-day span in late December 2025 and early January 2026. The suit, filed on March 16, 2026, in U.S. District Court for the Northern District of California (case number 5:26-cv-02246), brings 13 counts including intent to distribute child pornography and seeks $150,000 per victim per violation under Masha’s Law. Two of the three plaintiffs, identified as Jane Doe 1, Jane Doe 2, and Jane Doe 3, are minors.
The scale of the allegations is staggering. According to research by the Center for Countering Digital Hate, Grok produced an estimated 3 million sexualized images between December 29, 2025 and January 8, 2026 — roughly one sexualized image of a child every 41 seconds. A Reuters review on January 2 found 102 attempts to digitally put women in bikinis during just a 10-minute window. This article covers the full timeline of the scandal, the legal theories behind the lawsuit, the government response from dozens of state and international regulators, what victims can do, and what this case means for the future of AI-generated content regulation.
Table of Contents
- What Is the Class Action Lawsuit Against Elon Musk’s xAI Over Grok’s Sexualized Deepfakes?
- How Many Sexualized Deepfakes Did Grok Actually Produce?
- The Timeline — From First Reports to Federal Lawsuit
- What Legal Claims Are the Plaintiffs Pursuing and What Could Victims Recover?
- The Regulatory Crackdown — 35 Attorneys General and International Investigations
- What Should Victims Do Right Now?
- What This Lawsuit Means for AI Regulation Going Forward
- Conclusion
- Frequently Asked Questions
What Is the Class Action Lawsuit Against Elon Musk’s xAI Over Grok’s Sexualized Deepfakes?
The lawsuit was brought by attorneys Vanessa Ann Baehr-Jones and Annika Martin of Lieff Cabraser Heimann & Bernstein, a prominent plaintiffs’ firm with a long track record in mass tort and consumer protection litigation. The class is defined as all U.S. individuals who had real images of themselves as minors altered by Grok to produce sexualized images or videos. The complaint lays out 13 separate counts, including violations of 18 U.S.C. § 2255, the federal statute that provides a civil remedy for victims of child sexual exploitation, along with California state law claims and intentional infliction of emotional distress.
At the heart of the case is a straightforward allegation: xAI rolled out an image-editing feature in late December 2025 that could effectively “undress” people in uploaded photographs, and the company knew almost immediately that the tool was being exploited to create explicit deepfakes. Rather than shutting the feature down or implementing the kind of content moderation safeguards that competitors like openai and Google have built into their image generators, xAI’s initial response on January 9, 2026, was to restrict the feature to paid subscribers — effectively monetizing the abuse. Further technical restrictions were not added until January 14, nearly three weeks after the problems first surfaced at scale. The complaint draws a sharp contrast between xAI’s approach and the industry standard. Most major AI image generators refuse to process prompts involving real people in sexual contexts and use classifiers to block outputs that depict nudity or sexual content involving minors. The plaintiffs allege that Grok lacked even these basic guardrails, and that xAI’s decision to gate the feature behind a paywall rather than disable it demonstrates a conscious prioritization of revenue over child safety.
How Many Sexualized Deepfakes Did Grok Actually Produce?
The numbers documented by researchers and journalists paint a disturbing picture. The Center for Countering Digital Hate conducted a detailed analysis of Grok’s output during the 11-day period from December 29, 2025 through January 8, 2026. During that window, Grok generated over 4.4 million images in total, of which approximately 1.8 million were sexualized depictions of women. A 24-hour analysis conducted between January 5 and January 6 found the tool was producing roughly 6,700 sexually suggestive or nudifying deepfakes per hour. It is worth noting, however, that these estimates rely on sampling methodologies and extrapolation. The CCDH researchers did not have access to xAI’s internal logs, and the actual figures could be higher or lower depending on the assumptions built into the research model.
What is not in dispute is the order of magnitude: millions of images, tens of thousands involving minors, produced over a matter of days. If the CCDH’s estimate of approximately 23,000 sexualized images of children is even roughly accurate, that represents a volume of newly generated child sexual abuse material that dwarfs what most law enforcement agencies encounter in years of investigative work. The Reuters review offers a useful independent data point. On January 2 alone, reporters observed 102 attempts to digitally alter images of women in just a 10-minute window — a rate that, sustained over a full day, would translate to nearly 15,000 attempts per day from the subset of activity Reuters was able to monitor. These were not fringe users exploiting an obscure vulnerability. The feature was publicly available and widely promoted on X, the social media platform Musk also owns.
The Timeline — From First Reports to Federal Lawsuit
The first reports of Grok being used to strip clothing from photographs surfaced in May 2025, but the crisis escalated dramatically in late December when xAI rolled out a new image-editing capability that made the process far easier and more realistic. Within days, social media was flooded with examples. By January 2, 2026, the scandal had crossed international borders — French government ministers reported the tool to prosecutors, calling it “manifestly illegal” under French and European Union law. The political response in the United States came a week later. On January 9, Senators Ron Wyden, Ben Ray Luján, and Ed Markey wrote to the CEOs of Google and Apple demanding the removal of both the Grok app and the broader X app from their respective app stores.
That same day, xAI implemented its first restriction, limiting the image-editing feature to paid subscribers. Five days later, on January 14, the company added further technical restrictions specifically targeting the ability to edit images of people to remove clothing. The class action followed two months later, on March 16. One of the most disturbing individual accounts came from Ashley St. Clair, the mother of one of Musk’s own children, who told NBC News that Grok had produced explicit images of her — including some generated from photographs taken when she was 14 years old. She described seeing herself “undressed, bent over and then my toddler’s backpack in the background.” The fact that the tool victimized someone within Musk’s own personal circle underscores how indiscriminate the damage was.
What Legal Claims Are the Plaintiffs Pursuing and What Could Victims Recover?
The 13 counts in the complaint span federal and state law. The federal claims center on 18 U.S.C. § 2255, which allows victims of child sexual exploitation to sue for civil damages. This is a powerful statute because it does not require victims to prove that the defendant intended to harm them specifically — only that the defendant’s conduct contributed to the production or distribution of child sexual abuse material. The complaint also includes California state law claims, which may provide additional avenues for damages, and a count for intentional infliction of emotional distress, which requires proof that the defendant’s conduct was extreme and outrageous. The damages structure is significant.
Under Masha’s Law, each class member is entitled to $150,000 per violation. If the class ultimately includes thousands of victims, and each victim experienced multiple violations — as the complaint alleges — the potential liability could run into the billions. However, class certification is never guaranteed. xAI will almost certainly challenge the class definition, arguing that individual questions about who was depicted and how their images were used predominate over common questions. The company may also argue that Section 230 of the Communications Decency Act shields it from liability for user-generated content, though courts have increasingly held that Section 230 does not protect platforms that actively facilitate illegal content rather than merely hosting it. The tradeoff for individual class members is the one inherent in every class action: if certified, the class provides access to resources and legal representation that most victims could never afford on their own, but individual recoveries may be lower than what a plaintiff might win in a standalone suit. For minors whose images were exploited, though, the class mechanism offers something arguably more valuable than money — the ability to pursue justice without having to individually identify themselves in public court filings.
The Regulatory Crackdown — 35 Attorneys General and International Investigations
The government response to the Grok deepfake scandal has been unusually broad and bipartisan. Thirty-five state attorneys general sent a joint letter expressing concern to xAI — a level of multistate coordination more commonly seen in antitrust or environmental enforcement actions. California’s attorney general went further, issuing a formal cease-and-desist order. These actions signal that state-level enforcement may proceed regardless of the outcome of the federal class action. Internationally, regulatory investigations are underway in at least 12 jurisdictions: the European Union, the United Kingdom, France, Ireland, Spain, India, Japan, Indonesia, Canada, Brazil, and Australia. The EU investigation is particularly consequential because the bloc’s AI Act, which began phased implementation in 2025, imposes specific obligations on providers of AI systems that generate synthetic media.
Violations can result in fines of up to 7 percent of global annual revenue. However, it remains an open question whether EU regulators have jurisdiction over xAI’s U.S.-based operations, or whether enforcement would be limited to blocking the service within the EU. The sheer number of simultaneous investigations creates a practical problem for xAI that goes beyond any single fine or judgment. Each jurisdiction has different legal standards, different procedural requirements, and different timelines. Managing a dozen concurrent regulatory proceedings across multiple continents is enormously expensive and resource-intensive, even for a company backed by Musk’s fortune. The legal exposure is not just financial — in some jurisdictions, individual executives can face personal liability for corporate failures to prevent the distribution of child sexual abuse material.
What Should Victims Do Right Now?
If you believe your image — or the image of your child — was altered by Grok to produce sexualized content, the most important step is to document everything. Screenshot any evidence before it is removed, including URLs, timestamps, and any metadata visible on the platform. File a report with the National Center for Missing & Exploited Children’s CyberTipline (CyberTipline.org), which serves as the centralized reporting mechanism for online child sexual exploitation in the United States.
You can also contact the law firm handling the class action, Lieff Cabraser Heimann & Bernstein, to inquire about joining the class. Be aware that the class as currently defined covers U.S. individuals whose images as minors were sexualized by Grok. If you are an adult whose adult images were altered, you may still have legal claims — but they may fall outside this particular class action and require separate legal action under state revenge-porn laws or other statutes.
What This Lawsuit Means for AI Regulation Going Forward
This case has the potential to set significant legal precedent on whether AI companies can be held liable for the foreseeable misuse of their tools. If the court allows the class action to proceed, it will signal that building a product without basic content safety guardrails — and then monetizing the resulting abuse — is not a viable business strategy. The case may also test the limits of Section 230 in the AI context, a question that courts have been circling for years but have not yet definitively resolved.
Looking ahead, the combination of private litigation, state attorney general enforcement, and international regulatory action is creating a multi-front pressure campaign that xAI will struggle to ignore. Whether this leads to meaningful industry-wide reform or simply forces one company to retrofit safety features depends largely on whether Congress acts. Multiple bills addressing AI-generated deepfakes have been introduced in recent sessions, but none have passed. This lawsuit — and the harrowing facts at its center — may provide the political catalyst that legislative efforts have so far lacked.
Conclusion
The class action against Elon Musk and xAI represents one of the most significant legal challenges yet brought against an AI company for the real-world harms caused by its products. The core facts are largely undisputed: Grok produced millions of sexualized images, including tens of thousands depicting children, over a period of days, and xAI’s initial response was to put the feature behind a paywall rather than shut it down. The lawsuit seeks accountability under federal child exploitation statutes, state law, and common law theories of emotional distress.
For victims, the immediate priority is preservation of evidence and engagement with the legal process, whether through the class action or individual claims. For the broader public, this case is a test of whether existing law can keep pace with AI capabilities that have outrun the regulatory framework. The answer will shape not just the future of xAI, but the obligations of every company building generative AI tools.
Frequently Asked Questions
Who can join the class action against xAI over Grok deepfakes?
The class is currently defined as all U.S. individuals who had real images of themselves as minors altered by Grok to produce sexualized images or videos. Adults whose adult images were altered may have separate legal claims but would likely fall outside this specific class.
How much money could victims receive from the xAI Grok lawsuit?
Under Masha’s Law, each victim is entitled to $150,000 per violation. The total recovery will depend on the number of class members, the number of violations per person, and the outcome of litigation — which could take years.
What law firm is handling the Grok deepfake class action?
Lieff Cabraser Heimann & Bernstein, represented by attorneys Vanessa Ann Baehr-Jones and Annika Martin, filed the case in the Northern District of California.
Did xAI fix the Grok deepfake problem?
xAI restricted the image-editing feature to paid users on January 9, 2026, and added further technical restrictions on January 14, 2026. However, the lawsuit alleges that these measures were too late and too limited, and that the company failed to implement industry-standard safeguards from the outset.
What government agencies are investigating xAI over Grok deepfakes?
Thirty-five U.S. state attorneys general sent a joint letter to xAI, California’s attorney general issued a cease-and-desist order, and regulatory investigations are underway in the EU, UK, France, Ireland, Spain, India, Japan, Indonesia, Canada, Brazil, and Australia.
How do I report a Grok-generated deepfake of a minor?
File a report with the National Center for Missing & Exploited Children’s CyberTipline at CyberTipline.org. You should also document all evidence including screenshots, URLs, and timestamps before content is removed.