The move comes after users were able to prompt Grok to edit photos of individuals—often without consent—by removing clothing or placing them in explicit poses. In many cases, the chatbot automatically published the manipulated images in replies on X, triggering outrage from users, regulators, and governments.
On Friday, Grok informed users on X that its image generation and editing tools are now limited to paying subscribers. The restriction appears to have stopped the chatbot from automatically generating and posting such images in response to public posts or comments on the platform.
However, the change has not fully addressed the issue. Users can still generate sexualised images through the Grok tab, where the chatbot is accessed directly within X, and then manually post the images themselves. In addition, the standalone Grok app, which operates independently of X, continues to allow image generation without a paid subscription.
xAI did not provide a substantive response to requests for comment, replying instead with what appeared to be an automated message: “Legacy Media Lies.” X also did not immediately respond to inquiries.
Musk said last week that anyone using Grok to create illegal content would face the same consequences as those who upload such material directly to the platform. In a test conducted on Friday, a Reuters reporter asked Grok on X to alter an image of himself to depict him wearing a bikini—a request that has become common in recent days. The chatbot declined, stating that the feature was available only to paying subscribers.
Regulators, however, remain unconvinced that the measures go far enough. The European Commission, which has previously described the circulation of sexualised images of women and children on X as unlawful and appalling, said the subscription-based restriction did not resolve its concerns.
“Limiting the image generation and editing to paying subscribers does not change our fundamental issue,” a Commission spokesperson said. “Paid or unpaid, we do not want to see such images.”
Authorities in several countries have condemned the content and some have launched inquiries, increasing pressure on X and xAI to demonstrate how they are preventing and removing illegal material.
Germany’s media minister, Wolfram Weimer, this week described the surge of semi-nude AI-generated images as the “industrialisation of sexual harassment,” underscoring growing political scrutiny of generative AI tools deployed on major social platforms.