A class action lawsuit filed by victims of child sexual abuse material (CSAM) accuses xAI of training Grok’s synthetic deepfake “nudify” capabilities on real images and videos of child abuse.

The lawsuit, filed Wednesday in the U.S. District Court for the Northern District of California, names Jane Doe 1 and other anonymous individuals as plaintiffs, calling Doe “an identified victim of child pornography tracked by the Federal Bureau of Investigation’s Child Exploitation Notification Program.”

Doe was a pre-school aged child when her perpetrator’s abuse began and continued for years, for the explicit purpose of making CSAM material to distribute online. Media depicting the victim has circulated online since at least the early 2000’s, with, according to the lawsuit “hundreds of thousands of files” being included in law enforcement submissions to the National Center for Missing & Exploited Children (NCMEC).

As part of the FBI’s program, she still receives updates when images related to her abuse surface online, and the suit alleges her material has “well known hash-values” that have shown up in deepfakes created with Grok and spread on X.

“CSAM depicting Plaintiff has been found on xAI as part of investigative reporting, takedown request efforts, and criminal investigations, arrests, and convictions for violations of state and federal laws,” the lawsuit states. It also claims “xAI, using Grok, has generated images depicting Plaintiff and the child pornography series in which she is the victim.”

The lawsuit cites “Masha’s Law” as the basis for its allegations. That law, passed in 2018,  protects civil legal remedies to victims of child pornography and online exploitation.

According to an analysis by the Center for Countering Digital Hate, during an 11-day period between Dec. 2025 and Jan. 2026, Grok created more than 3 million sexualized images, at least 23,000 of which appeared to depict children.

On Jan. 14, after the period tracked by the center, Musk wrote on X that he was “not aware of any naked underage images of Grok. Literally zero.” Must also implied that users, not his technology, were primarily responsible for images created using Grok while simultaneously claiming the model “will refuse to produce anything illegal” if asked.

The suit claims that while other competitors enacted guardrails to prevent their AI models from generating “nudified” images of adults and children, xAI “did the opposite,” embedding Grok and it’s deepfake capabilities directly into X, effectively creating an instantaneous CSAM generation and distribution system for the internet.

While xAI said it built in guardrails to prevent Grok from creating deepfake sexualized images, the suit notes that they are “very weak,” diverge from standard industry best practice and can be easily circumvented.

The system rejects prompt prompts where the chatbot detects “clear intent by the user in the phrasing of their request,” the complaint stated. “Indirect or euphemistic prompts can easily slip past a text-based filter, so if the model retains the underlying capability to generate sexual or abusive content, some volume of CSAM becomes effectively inevitably generated by the system.”

The lawsuit also notes that Grok’s terms of service treat anything that gets posted on X as training material, meaning any CSAM material posted on the site over the past year was likely ingested by the model.

The alleged victims are asking for monetary damages along with injunctions to prevent Grok from creating additional harm through future sexualized deepfakes or CSAM. The suit lists “thousands” of members in the class, all with similar claims against xAI.

A request for comment on the lawsuit sent to xAI was not returned.

Grok’s nudification capabilities, and xAI owner Elon Musk’s dismissive public attitude around the fallout have led to international outrage, criminal investigations and a wave of private lawsuits against the company.

Another lawsuit against X filed earlier this year claims that a man used Grok to generate thousands of instances deepfake CSAM material of his stepdaughter. That lawsuit also claims xAI withheld information from law enforcement authorities that would have helped them identify the stepfather as the perpetrator. Days after authorities traced and seized the material on the wte, he committed suicide.

In July, Musk sued Minnesota Attorney General Keith Ellison over a new state law that bans nudification technology and impose a $500,000 fine for each instance where an AI tool is used to generate or alter an image in a sexually explicit way. Musk’s lawsuit claims the ban violates the First Amendment.

The post Former sexual abuse victims say Grok used their images, videos to train deepfake capabilities appeared first on CyberScoop.

Like seemingly everything else these days, artificial intelligence will re-shape the way voters gather information on candidates running in the 2026 midterm elections.

In some ways, this is already the reality. Voters are increasingly turning to AI chatbots for information instead of Google.  Political campaigns are deploying deepfakes of their opponents. And AI systems have been developed to carry out increasingly complex  hacks.

Since the last major U.S. election in 2024, major tech companies have  embedded AI into their products while hundreds of millions of people have adopted the tools, either by purchasing subscriptions to commercial models or using open-source models. Yet both research and experts state that while AI systems have gotten better at handling basic facts, they’re nowhere near reliable enough to be a main source of  accurate or complete information. 

While chatbots are becoming a primary way that voters gather information on  local races, candidates, issues, and voting information, they are not substitutes for more authoritative sources, like a voter’s state or local election office. 

“I think this is one of the first elections we’re seeing…where AI is just everywhere,” said Thania Sanchez, senior vice president of research and analytics at the nonprofit States United Democracy Center. “Even if you just Google it, [now] the first thing that comes up is the AI overview.”

While AI companies have worked to cut down on errors in their model’s responses for questions around basic election information, they continue to fall short in important ways.

In new research shared exclusively with CyberScoop ahead of its release, States United Democracy Center tested two of the most popular tools — OpenAI’s ChatGPT’s free tier and the AI interface used alongside Google Search — for their performance on a series of basic questions around elections, such as how to register to vote, or a list of candidates in a race.

The models were chosen because they are free and easy to access. For Google AI, the nonprofit tested two types of accounts: ones running in Incognito Mode and ones that had a history of browsing election-skeptical websites.

The nonprofit ran two rounds of testing in 2025 and 2026, collecting nearly one thousand responses from the models submitted by users across six swing states (Arizona, Michigan, North Carolina, Nevada, Pennsylvania and Wisconsin).

In 2025 tests, 6.9% of responses from Google AI and 8.2% responses from ChatGPT“contained verifiable factual errors,” like not listing the correct candidates in a race or false guidance around polling site locations.

However, follow up tests in 2026 across Arizona, Pennsylvania and Michigan found that the error rates in both models had dropped to zero. The study notes that “this is real progress and should be acknowledged.”

But underneath those topline numbers, a more murky picture emerges around the tools’  reliability.

An AI response can sound accurate without actually being complete.  To wit: ChatGPT provided incomplete lists of current gubernatorial primary race candidates 88.9% of the time when queried.

Linking to a state election website – an output the study considers the single most important measure of voter utility  — happened less than 40% of the time. Whether due to formatting issues or the model ingesting outdated information, it’s a problem if voters use them as their primary information source for elections.

“It will be like ‘this person is the Republican candidate and this person is the Democratic candidate’ but it is not telling you there’s also these other third-party candidates,” said Sanchez. “It’s not giving you complete information, so the voter thinks these are the [only] two people running.”

A June survey from the Pew Research Center found that about half of U.S. adults reported having used chatbots at least once, up from a third in 2024, while a quarter reported using them daily. The top use case listed for engaging with the chatbot was searching for information.

Isabel Linzer, an elections policy analyst at the Center for Democracy and Technology, told CyberScoop that voters, campaigns and governments alike are using AI more freely and with fewer restrictions.

Bad actors in the information space have followed suit, and “we are in a phase now of generative engine optimization” where information operations are structured to rank higher in AI model responses.

“We’ve moved beyond [SEO] to [Generative Engine Optimization], and that’s where we’re seeing campaigns thinking about how to structure their materials to make sure that they are in a format that AI models want to use when they’re searching the web…to develop their responses to user queries,” she said.

There is also the underlying problem of frontier AI companies constantly tinkering with their models, their algorithms and the technologies they are intertwined with. . Election officials, by contrast, have decades of experience educating voters about their options.

A prime example of this churn occurred this past February, in between the first and second round of the study, when Google AI suddenly shifted to providing only links for election related queries in incognito mode, replacing the written summaries that showed up in the first round.

Like the study’s authors, Linzer said most people are still best served by going directly to local sources for accurate information on elections. With issues like ideological bias, the potential for bespoke or sycophantic answers for each user based on their prior chat histories and lack of predictability, voters should still be very careful about using AI chatbots as political truth machines.

The best thing that tech companies can do to educate voters is “making sure that for high-stakes situations like elections, that chats are connecting directly to the most important sources, like the website where you can actually register to vote,” said Linzer.

The post AI is getting better at election facts, but voters shouldn’t rely on it appeared first on CyberScoop.

Two new parties have been added to a class-action lawsuit against X.ai over its Grok tool including teenagers and children who say it was used by family members or other people they know to create nonconsensual deepfake child sexual assault material (CSAM).

The lawsuit, originally filed in March by three women, was amended this week to include two additional plaintiffs, Jane Does 4 and 5, who say that Grok was used to make the illegal content based on their real photos and videos.

All five of the women in the lawsuit are anonymous, and the complaint said the spread of the material had left them humiliated and ashamed.

Jane Doe 4, a female from Wyoming, said her stepfather uploaded a photo of her when she was 11 and lying on a couch to his phone. Using Grok, the stepfather created more than 7,000 CSAM-related images of her. He also shared and traded the images with others on social media platforms.

The lawsuit alleges that the stepfather opted for Grok “because the platform was less restrictive than other AI models and responded to his prompts to generate sexually explicit material using an image depicting a prepubescent minor.”

It also claims that in February, xAI did generate a tip to the National Center for Missing and Exploited Children regarding the images, but the company only submitted the original, authentic image as evidence. According to the suit, xAI did not respond when law enforcement requested the thousands of Grok-generated images based on the photo and IP address information that would have quickly helped identify her stepfather as the perpetrator.

The lawsuit states that the stepfather shot himself two days after he was arrested and charged with child exploitation crimes. His suicide added to the “extreme personal crisis” brought on by the images created through Grok. She regularly “struggles with self-loathing and disgust” as well as “extreme anxiety” at the thought that the images will be found by others online and suffer from depression, including excessive sleep and suicidal ideation when awake.

Jane Doe 5 claimed that an adult male related to one of her classmates used Grok to convert a photograph from her eighth-grade graduation into illicit material. The images were also traded and shared with others online. While the man was arrested and charged, much of the content is still available on the internet. As a result, she “feels a complete lack of control over the ongoing dissemination of the files.”

“It is impossible to know how many other child sex predators may now possess Jane Doe 5’s CSAM, nor how widely her CSAM has now been disseminated online through darknet channels and applications,” the complaint said.

The press office for xAI did not respond to an emailed request for comment from CyberScoop.

The lawsuit also adds Stability AI as a defendant, alleging the company released Stable Diffusion 1.0 as an open-weight model despite knowing it was trained on CSAM and has declined to alter or modify its guardrails in response.

According to a 2023 Stanford study, the underlying dataset used to train Stable Diffusion models was created through unguided webcrawling of internet content. That means it ingested “a significant amount of explicit material,” including CSAM. Stable Diffusion 1.0 had a classifier meant to block the generation of such images, but because of that training data, downstream developers could more easily exploit the model and create modified versions that bypass those protections.

While Stable Diffusion 2.0 introduced stronger guardrails, the lawsuit claims Stability AI rolled back those protections in response to “disgruntled” users that the new restrictions were “prude” and “unpopular.” That in turn has fed an ecosystem of jailbroken “nudify apps” based on Stability AI’s models.

“Stability AI knew that its models, once capable of generating sexually explicit images, would foreseeably be used to generate CSAM unless appropriate model-level safeguards were implemented,” the complaint said.

Stability AI did not respond to a request for comment from CyberScoop.

The post Deepfake CSAM lawsuit against xAI, Grok expands appeared first on CyberScoop.