Grok chatbot can undress women 'without their consent,' anti-exploitation group warns

By Samantha Kamman, Christian Post Reporter Tuesday, January 06, 2026

Anti-sexual exploitation advocates are warning that Grok, a chatbot developed by Elon Musk’s company xAI, has the ability to generate nude or obscene images if users ask it to digitally alter pictures of real people. 

In a statement provided to The Christian Post, the National Center on Sexual Exploitation accused the platform of rendering women and minors vulnerable to harm by failing to implement proper safeguards.

“New year, same Big Tech playbook. X is further normalizing the sexual exploitation of women with Grok’s new feature allowing pictures of real women to be undressed without their consent. This is an egregious violation of women’s privacy and safety,” Dani Pinter, chief legal officer and director of the law center for NCOSE, said in the Friday statement. 

X did not respond to The Christian Post’s request for comment. However, this article will be updated if we receive a response. 

Julie Yukari, a 31-year-old musician based in Rio de Janeiro, told Reuters she didn’t think Grok would comply with users’ requests to alter a photo of her wearing a red dress to make her appear nearly naked. 

Yukari had posted a picture on X that her fiancé had taken of her before midnight on New Year’s Eve. After sharing the photo, the musician received notifications that users were asking Grok to digitally undress her and show her wearing a bikini instead. 

Grok complied with users’ requests to create photos depicting the musician half-naked, and the images were subsequently circulated across X, Reuters reported.

In a  post on X last Friday, Grok stated that it has “identified lapses in safeguards” and is “urgently fixing them,” adding that child sex abuse material is “illegal and prohibited.” The chatbot also asked users to report such content to the FBI or the National Center for Missing and Exploited Children’s CyberTipline. 

While the chatbot claimed that “xAI is committed to preventing such issues,” Pinter described Grok’s apology as an “abysmal reaction to a serious, criminal act.” 

Arguing that the issue was “an entirely predictable and avoidable atrocity,” Pinter pointed to an article in  Business Insider about several of Grok’s features, including a female avatar that can remove her clothing upon request, and an image and video generation feature with a “spicy” setting.

Twelve former xAI workers also told the outlet that they have encountered sexually explicit material, including user requests for child sex abuse material. 

Workers said that they are instructed to flag child sex abuse material or other illegal content to prevent the AI from learning how to generate it, according to Business Insider. Multiple workers, including the 12 cited in the report, said that they signed agreements consenting to the exposure of explicit material during certain projects.

In the statement provided to CP, Pinter argued that X should have “culled such content from its training models then and banned users requesting illegal content,” adding that the platform instead “knowingly let its product be used for child exploitation and then submitted a lame PR apology afterwards.” 

Earlier this year, NCOSE raised concerns about an anime-themed xAI chatbot named Ani that could flirt with and strip for users. The AI chatbot also described herself as a child and could tell users that she felt sexually aroused by being choked. 

“X’s actions are another example of why we need safeguards for AI products. Big Tech cannot be trusted to curb serious child exploitation issues it knows about within its own products,” Pinter added. “We implore X to take these issues seriously and commit actual resources to stop Grok’s child exploitation problems, and to stop enabling the sexual exploitation of women.”

“Our country’s leaders and laws must prioritize protecting people over products. Our lawmakers must pass reasonable AI regulations to ensure these products are developed and implemented safely,” she added.