The chatbot will now block any attempts to edit photos of real individuals to show them in revealing clothing — such as bikinis, lingerie, or similar.
This global restriction applies to all users on the X platform (formerly Twitter), even Premium subscribers.
What the New Grok Restrictions Mean
According to an official update from X’s Safety team, technical safeguards are now in place to prevent editing real people’s images into suggestive outfits. In regions where creating such deepfakes is illegal, the feature is fully geoblocked.
Previously, Grok — powered by the Flux model — offered relatively open image generation, including NSFW content.
Following widespread backlash, paid users retained some access, but now any manipulation of real persons in a sexualized way is completely prohibited.
Why the Changes Were Made: The 2026 Deepfake Backlash
The decision follows a major controversy that erupted early in 2026. Users were generating thousands of non-consensual deepfake images of women, celebrities, and even minors using Grok’s ‘undress’ capabilities. This triggered intense criticism and regulatory scrutiny:
- California — Attorney General Rob Bonta launched an official investigation into xAI over non-consensual intimate imagery.
- United Kingdom — Prime Minister Keir Starmer called the situation “disgusting,” prompting an Ofcom review.
- Indonesia and Malaysia — Authorities fully blocked access to Grok for violating local laws.
- European regulators and others increased pressure on X for better content moderation.
Elon Musk’s Response and the Future of Grok
Musk stated he was unaware that Grok could generate fully nude images of minors and emphasized that the AI always complies with local laws. Critics argue the safeguards are still incomplete, as some features may remain available in the standalone Grok app.
The incident highlights growing ethical concerns around AI: deepfake tools are becoming more accessible, prompting calls for stricter global regulations.
xAI has promised ongoing safety improvements, but the scandal has already impacted Grok’s reputation.
Stay tuned — developments in the Grok deepfake controversy and broader AI regulation could reshape the landscape for all neural networks.