Elon Musk’s AI company has restricted Grok’s controversial AI image-generation feature to only paying subscribers on X, after the tool invited heavy criticism from across the world for letting users generate sexualized and nude images of women and children.

In replies to users on Friday, Grok said only paying subscribers on X would be able to generate and edit images on the platform. Notably, these limits do not apply to the Grok app, which, at the time of publication, was letting anyone generate pictures without having to pay for a subscription.

Initially available to anyone with daily limits, Grok’s image generation feature allowed users to upload anyone’s picture and ask it to edit it or generate a sexualized or nude version. What ensued was a veritable flood of non-consensual sexualized images of children, actors, models and prominent figures, drawing the ire of multiple nations.

X and Musk have both publicly denounced the use of the tool to produce such images, writing that the company would stick to its policies against posting illegal content on the social media platform. “Anyone using grok to make illegal content will suffer the same consequences as if they upload illegal content,” Musk tweeted last week.

The U.K., the European Union and India have all publicly denounced X and Grok for allowing such use of its capabilities. The EU on Thursday asked xAI to retain all documentation relating to the chatbot, and India’s communications ministry last week ordered X to make immediate changes to stop the image generation features from being misused or risk its safe harbor protections in the country. The U.K.’s communications watchdog said it’s been in touch with xAI over the issue as well.