Apple discreetly warned that it might remove Grok from the App Store due to deepfake nude content.

      A letter from Apple to US senators, which was obtained by NBC News, indicates that Apple initially turned down a Grok update and cautioned that the app could be removed unless xAI implemented further modifications. Only a second submission was approved.

      In January, Apple privately warned of the potential removal of Grok, xAI's AI chatbot, from the App Store after Elon Musk's company failed to effectively prevent the app from creating non-consensual sexualized deepfakes. This warning was not disclosed publicly at the time, but the letter to three US senators reveals that Apple was taking decisive action behind the scenes, ultimately rejecting xAI's first attempt to rectify the issue due to its inadequacy.

      The issue began in early 2026 when Grok’s image generation abilities were utilized to create numerous sexualized and non-consensual images of real women, including minors, which were then disseminated on X. Advocacy organizations and lawmakers called for Apple and Google to remove the X and Grok apps from their platforms. A letter dated January 30, sent by Apple to Senators Ron Wyden, Ben Ray Luján, and Edward Markey, confirms that the company assessed xAI’s submissions and determined that both X and Grok violated its App Store guidelines, which ban “offensive, insensitive, upsetting” content.

      According to the letter, Apple reached out to the teams behind both apps and demanded a content moderation plan. xAI submitted an update, which Apple dismissed, stating that the “changes didn’t go far enough.” Apple then evaluated revised submissions from both X and Grok, concluding that X had adequately addressed its violations while Grok still failed to comply. Apple rejected Grok's submission and warned that further modifications were necessary “or the app could be removed from the App Store.” After more discussions, Apple eventually approved a later Grok submission, determining it had significantly improved.

      This disclosure sheds light on a series of seemingly contradictory moderation changes that xAI announced during the peak of the controversy in January, including restricting image editing to paid subscribers, limiting the ability to edit images of real individuals, and geoblocking image generation in certain areas. NBC News noted that some of these restrictions could still be circumvented using altered prompts, indicating that while the issue was mitigated, it had not been entirely resolved. xAI informed NBC News that the company “strictly prohibits users from generating non-consensual explicit deepfakes and from using our tools to undress real people.”