Class action suit against AI makers over deepfake child sexual abuse material expands
In this photo illustration, the Grok website is seen through a magnifying glass on a computer screen on February 12, 2026. Grok is the AI chatbot built by Elon Musk’s SpaceXAI, formerly known as xAI. A class action lawsuit filed against SpaceXAI and Stability AI alleges the companies’ AI tools were used to make sexually explicit images of children.
Pablo Vera/AFP via Getty Images
hide caption
toggle caption
Pablo Vera/AFP via Getty Images
New plaintiffs in a class action lawsuit against Elon Musk’s AI company, SpaceXAI, allege that the company’s image-generation models were used to create child sexual abuse material and that the company failed to adequately share information about an alleged perpetrator with authorities.
Two plaintiffs, one in Wyoming and the other in Wisconsin, joined the lawsuit filed by three Tennessee teenagers earlier this year, according to an amended complaint filed on Tuesday. All the plaintiffs are referred to as Jane Does. The suit was also expanded to include Stability AI, the company behind the Stable Diffusion image generator, as a defendant.
The suit alleges that the perpetrators, including a plaintiff’s male friend and another plaintiff’s stepfather, used the companies’ AI models to alter photos taken when the plaintiffs were underage to make child sexual abuse material (CSAM). The five plaintiffs accuse the companies of producing CSAM, benefiting from sex trafficking ventures, negligence, defective product design and creating a public nuisance.
“Public nuisance means that this is not just something that is problematic for our clients … this is something that is a scourge on society,” said Annika Martin, an attorney for all five plaintiffs. “We want to put these guardrails in place so that we do not cause this harm across an entire generation of children.”
The plaintiffs are asking the AI companies to install more effective guardrails to prevent the creation of exploitative and abusive imagery as well as seeking monetary compensation.
The case of Jane Doe 4
While SpaceXAI, like other internet companies, is required by law to report suspected child sexual exploitation, including CSAM, to the nonprofit National Center for Missing and Exploited Children (NCMEC), the lawsuit alleges that the company failed to adequately do so.
SpaceXAI was previously known as xAI. The company rebranded this month following its merger with Musk’s SpaceX earlier this year.
According to the complaint, the stepfather of Jane Doe 4, a woman in her 20s in Wyoming, used SpaceXAI’s chatbot, Grok, to generate about 7,000 sexually explicit images and videos from a single photograph taken when Jane Doe 4 was about 11. The AI-generated images included depictions of her nude, performing sexual acts on men, including her stepfather. Some images also included explicit captions.
SpaceXAI only sent one tip regarding Jane Doe 4 to NCMEC in February, when the stepfather asked the model to generate an image depicting the girl being raped by multiple men, according to the complaint. The company did not include any of the abusive images with its report, and also failed to share the alleged perpetrator’s IP address with NCMEC and law enforcement, even after officials requested it multiple times, according to the complaint.
Many electronic service providers submit a large volume of reports of suspicious activity to NCMEC, but do not include “sufficient or actionable information” because the law did not require it, the center wrote in a report in March.
When law enforcement eventually tracked down and investigated the stepfather, they found that he, like the other perpetrators mentioned in the lawsuit, traded the sexually explicit images with others online, according to the complaint. Two days after law enforcement searched his digital devices and charged him with child exploitation offenses, he died by suicide.
“Jane Doe 4 entered a period of extreme personal crisis,” the complaint said. “She had to grapple with the trauma of her own sexual exploitation while at the same time assisting her mother in navigating the loss of her stepfather … Her family was torn apart, and her life became a nightmare.”
Law enforcement told Jane Doe 4 that her stepfather had used Grok because it was more responsive to his prompts than other AI models, according to Martin, the lawyer representing the plaintiffs in the class action lawsuit.
NCMEC declined to comment on the case and referred NPR to law enforcement. The Wyoming Division of Criminal Investigation, whose task force on internet crimes against children was identified as one of the law enforcement agencies involved in Jane Doe 4’s case in the complaint, did not respond to a request for comment. SpaceXAI did not respond to a request for comment.
Apps that can alter photographs to make intimate images and pornography have existed for years in the shadows of the internet. But the creation of nonconsensual intimate images became much more mainstream in 2025 when major AI companies including Google, OpenAI and xAI updated their image generation tools in a way that allowed users to strip people down to bikinis.
From late 2025 into early this year, people used xAI’s tools to make a large number of altered photos showing women and children stripped down to bikinis or even less. As a result, countries including Indonesia, Malaysia and the UK launched investigations and imposed temporary bans on xAI. Currently, according to NPR’s testing, chatbots created by Google and OpenAI refuse to respond to the prompt “put her in a bikini.” A free version of Grok showed an error in loading images after receiving the same prompt.
The Stability AI logo is being displayed on a smartphone in this photo illustration on June 10, 2024. A class action lawsuit filed against SpaceXAI and Stability AI alleges the companies’ AI tools were used to make sexually explicit images of children.
Jonathan Raa/NurPhoto via Getty Images
hide caption
toggle caption
Jonathan Raa/NurPhoto via Getty Images
Stability AI is a new defendant
Stability AI, which makes the image generation tool Stable Diffusion, was added as a new defendant in connection with the claims filed by the three initial teenage plaintiffs in Tennessee. The complaint was updated to add that “The application on the perpetrator’s phone used to create the AI CSAM of Plaintiffs relied on Stability AI’s image-producing tools.”
Unlike proprietary models such as those made by SpaceXAI, OpenAI and Anthropic, Stability AI’s models are open-weight, which means it’s much easier for users to remove restrictions that model makers have put in place. What’s more, the model maker generally doesn’t have insight into user queries. Nonetheless, the lawsuit alleges Stability AI could have restricted its models’ ability to generate CSAM but chose not to in its more recent releases.
The Verge reported in November 2022 that Stability AI said it filtered out not-safe-for-work (NSFW) content from its training data to restrict the Stable Diffusion Version 2 model’s ability to generate CSAM. This decision drew backlash from users, including one who decried the update as “censorship,” The Verge reported.
A more recent version of Stability AI’s model was frequently used to generate NSFW content and generated a higher proportion of such content compared with outputs from Stable Diffusion Version 2, according to a research paper by scholars from the University of Wisconsin-Madison, the Massachusetts Institute of Technology and other institutions, which has not yet been peer-reviewed. Stable Diffusion Version 2 was also much less popular than the Stability AI models that came before and after it, the researchers found.
“When they realized that nobody wanted to use their model when they put those guardrails in place, they rolled those guardrails right back so that everybody would want to use their model again,” Martin said. “They knew exactly how to rein this behavior in and they chose not to for profit.”
Stability AI did not respond to a request for comment.
- Admin
- Scarborough says 'hypocritical' Republicans should 'shut up' over criticism of Dems' early support for Platner July 10, 2026
- Serial killer Joel Rifkin’s LI murder home finds new owners after 3 months for sale: ‘They’re so excited’ July 10, 2026
- Smithsonian chief emphasizes 'accuracy and integrity' after White House report July 10, 2026
- Sophie Cunningham further humiliates her finger-wag victim, a shameful Bryce Harper video & MLB has gone soft! July 10, 2026
- Taylor Strecker teases ‘House of Stassi’ drama and more with VRT July 10, 2026