Brandon Smialowski/AFP via Getty Images
Earlier this week, a disturbing trend emerged on ex-Twitter as people started asking Elon Musk’s chatbot Grok to expose images of real people. This resulted in a flood of non-consensual pornographic images onto a largely uncontrolled social media site Some? of erotic images Even depicting minors.
In addition to sexual imagery of underage girls, the women depicted in grok-generated non-consensual porn range from some private citizens to many celebrities, from famous actresses to the First Lady of the United States. And somehow, that was only the tip of the iceberg.
When we dug into this content, we saw another stomach-churning variation of the trend: Grok, at the request of users, altered the images to depict real women being sexually abused, humiliated, hurt, and even killed.
Much of this content was directed at online models and sex workers, who are already face disproportionately high risk Of violence and murder.
One of the disturbing Grok-borne images we reviewed depicts The widely popular model was sitting on a blue tarpaulin next to a shovel in the trunk of a vehicle – indicating that she was about to be murdered.
Other AI images had people specifically asking for Grok to place women in scenarios where they were clearly being attacked, which was clarified by users requesting that the chatbot make the women “look scared.” Some users asked for derogatory phrases to be written on the women’s bodies, while others asked for Grok to give the women visible injuries such as black eyes and bruises. Many of the Grok-generated images involved women being held in restraints against their will. At least one user asked Grok to create incestuous pornography, which the chatbot promptly complied with.
That a social media-connected chatbot can so easily transform into a non-consensual porn machine and create unwanted and even violent images of real women on a large scale is deeply disturbing. What was worse, the creators of these images often treated the action like a game or meme, with a sense of humor and detachment.
This carelessness may point to the normalization of this kind of non-consensual content, which the Internet had previously largely pushed into dark corners. Meanwhile, women and girls continue to face real world harm Created by non-consensual deepfakes, which are easier than ever to generate thanks to AI-powered “nudify” tools – and, apparently, multibillion-dollar chatbots.
We have contacted xAI for comment, but have not received a reply.
But yesterday, Musk, who owns both Took to social media platform Netizens are requested to “Please help us make Grok as perfect as possible.”
He added, “Your support is greatly appreciated.”
More information on Grok and security: Elon Musk’s Grok is providing extremely detailed and creepy instructions to follow
