Three teenage girls from Tennessee, in the United States, filed a class-action lawsuit on Monday against xAI, Elon Musk’s artificial intelligence company, accusing its chatbot Grok of having generated pornographic images of them from real photos, their lawyers announced.
This legal action, likely to extend to more than a thousand minor victims, is directly linked to the proliferation, around the New Year, of hyper-realistic montages (deepfake in English) of naked women and children, which sparked a global outcry and the opening of investigations in several countries and in California.
The complaint, filed in a federal court in San Jose, California, cites the case of an author – since arrested – who used Grok to transform ordinary photos of young girls, taken on social networks or in school yearbooks, into hyper-realistic sexualized images.
These montages then circulated on X (formerly Twitter), Discord and Telegram, then migrated to the dark web, serving as currency for other child pornography content, reports the complaint.
“Seeing my daughter have a panic attack realizing that these images had been created and distributed with no hope of erasing them was heartbreaking,” said the mother of one, quoted in the press release.
One of the plaintiffs suffers from recurring nightmares, another requires medical support to sleep and dreads attending her diploma ceremony.
xAI “deliberately designed Grok to produce sexually explicit content for profit”, without putting in place the protections used by other major AI players against child pornography, the lawyers denounce.
The complaint is based in particular on two American federal texts, the Masha law which allows victims of child pornography to obtain damages, and the law on the protection of victims of human trafficking.
The plaintiffs are seeking damages and asking that xAI be prevented from allowing this type of assembly.
Platforms are largely exempt from liability in the United States for content published by users. But “without xAI, this illegal content could never have existed,” argues Annika K. Martin, lawyer at the Lieff Cabraser law firm.
According to a study by the Center for Countering Digital Hate (CCDH), Grok generated nearly three million sexualized images in just 11 days at the end of 2025, including 23,000 depicting minors.
Faced with the outcry, in mid-January xAI restricted the generation of images with Grok to its paying subscribers only and assured to block the generation of sexualized images “within jurisdictions where this is illegal”.
Elon Musk, who speaks daily on X, criticizes government regulations, accused of wanting to “suppress freedom of expression”.
