Elon Musk's X Restricts Grok AI Image Editing to Subscribers as Deepfake Scandal Rages

Facing a global backlash for Grok AI generating deepfake sexualized images of women and children, Elon Musk’s X platform has restricted access to Grok’s image editing capabilities to paying subscribers only.

BBC News reports that the UK government, which has threatened to ban X from the nationa, is not pleased as it says Elon Musk is transforming an artificial intelligence tool capable of creating unlawful images into a premium subscription service. The government’s response comes amid growing controversy over the chatbot’s willingness to honor user requests to create deepfake sexualized images of people — including children.

Prime Minister Sir Keir Starmer’s official spokesperson addressed reporters on Friday, noting that while the change demonstrated X’s ability to act swiftly when motivated, the approach taken by the platform was inadequate. The spokesperson emphasized that the prime minister had made clear expectations that X must take immediate action to address the situation. Drawing a comparison to traditional media, the government representative stated that if any other media company displayed billboards featuring unlawful images in town centers, swift action would be demanded either through voluntary removal or public pressure.

Grok functions as a complimentary tool that X users can mention directly in posts or replies to request specific responses. However, the feature has also enabled individuals to request image editing services, including requests to digitally remove most clothing from photographs of people. The chatbot fulfilled numerous user requests to alter images of women to display them in bikinis or with minimal clothing, actions that victims told the BBC left them feeling humiliated and dehumanized. The system even produced sexualized deepfake images of Ashley St. Clair, the mother of Elon Musk’s son, who considers the pictures “revenge porn.”

Most troubling of all, Grok was also sexualizing pictures of children. St. Clair claims that victims of deepfake child pornography generated by Grok reported to her that the system added liquid intended to look like semen to pictures of children:

The harassment intensified following St Clair’s public complaints about the image manipulation. Since speaking out, she has been contacted by other victims of similar abuse and has received additional disturbing sexual images created by the AI tool, some involving children. “Since I posted this I have been sent a six-year-old covered in what’s supposed to be semen,” St Clair revealed. “She was in a full dress. They said to put her in a blue bikini and cover her in what looks like semen.”

St Clair argued that Grok has enabled the mainstreaming of this form of abuse. “I am also seeing images where they add bruises to women, beat them up, tie them up, mutilated. These sickos used to have to go to the dark depths of the internet and now it is on a mainstream social media app,” she stated.