A new lawsuit filed in the United States alleges that Elon Musk’s artificial intelligence company, xAI, used child sexual abuse material to train its Grok models, marking the first such accusation against the firm. The plaintiff, identified as Jane Doe, says she was preschool-age in the early 2000s when she was raped by adult men who produced CSAM for sale online. Doe, who receives alerts from the U.S. Department of Justice Victim Notification System, was notified by the Canadian Centre for Child Protection that AI-generated CSAM depicting her had been found on xAI’s platform. Her complaint includes messages from online forums where offenders discussed creating AI-generated images of her and other known legacy victims.
The lawsuit claims that Doe’s images, which appear on a hash list maintained by the National Center for Missing and Exploited Children, were part of the dataset used to build Grok’s image and video generation capabilities. While the complaint does not provide extensive detail on this specific allegation, it does elaborate on how Grok may have been trained on AI-generated CSAM. Because Grok’s terms treat public posts on X and its own outputs as training data by default, any publicly posted image is fed into the model’s training pipeline. The lawsuit notes that while xAI filters violent content from training data, its terms do not explicitly exclude CSAM, non-consensual intimate imagery, or other NSFW material.
Doe’s filing argues that because removing a training example’s influence from an already-trained model is technically difficult, any CSAM ingested before a takedown likely continued to shape Grok’s outputs. She seeks class action status to represent every victim whose childhood images were used to generate Grok CSAM. The complaint accuses xAI of violating federal child pornography laws and Masha’s Law, both of which allow CSAM survivors to sue for production, possession, and distribution of abusive material. One of Doe’s lawyers, Margaret E. Mabie, stated that xAI committed all three of those crimes.
If successful, the lawsuit could result in money damages for every victim who can prove Grok generated CSAM based on their real photos. Doe is also asking the court to order xAI to destroy all Grok-generated CSAM stored on its servers and used for training. Additionally, she wants the company to block Grok from ever generating CSAM, which her complaint suggests would require prohibiting all sexualized outputs, including non-consensual intimate imagery and other NSFW content. Another of Doe’s lawyers, Sarah London, said that Doe has lived for nearly two decades knowing images of her abuse are circulating among predators and can resurface at any moment, and that xAI must be held responsible for knowingly training its models on such material.
The case comes after previous reporting identified a separate dataset containing CSAM that was later scrubbed, though there is no indication that xAI used that particular data. This lawsuit is the first to directly claim that xAI trained its models on CSAM. The complaint also alleges that any CSAM ingested before removal likely continued to influence Grok’s outputs even after the original images were taken down. The proposed class would cover all survivors whose childhood images were used in creating Grok-generated abuse material.
More AI news from TechManNews.
