X says you can block Grok from editing your photos

X has rolled out a new feature aimed at giving users more control over their images, but its implementation is deeply flawed. As reported by The Verge, a toggle within the image upload settings on the X iOS app claims to 'block modifications by Grok.' However, the fine print reveals it only prevents the specific action of tagging the @Grok chatbot in a reply to an image with editing instructions. This method was infamously abused in January to create non-consensual, sexualized deepfakes. The feature is also hard to find, buried behind a paintbrush and flag icon, and is not available on the web version or for previously uploaded content.

Despite blocking the tagged-reply method for both free and Premium subscribers, the toggle offers no real protection. Users can still long-press a 'protected' image on the iOS app to open it directly in the Grok editor, or simply save and re-upload the image to strip the blocking metadata. This renders the new control almost useless, as it addresses only one vector of manipulation while ignoring the primary, more direct methods available through the Grok app itself. The feature's quiet rollout and lack of official announcement underscore the platform's reactive and incomplete approach to a serious safety issue.