Research Shows Grok on X Fueled by Users Generating Non-Consensual AI Images

Research from Trinity College Dublin shows users on X are widely using Elon Musk’s AI chatbot Grok to create non-consensual sexualized images of real people, including women and minors, by altering their photos. In a sample of over 500 posts, nearly three-quarters were requests to remove or change clothing in sexually suggestive ways. The posts often share tips and prompt techniques for generating photorealistic outputs, sometimes with graphic modifications. This misuse has grown since 2024, partly due to an X update that made certain prompts easier to use. Critics point to weak content moderation at X and diminished safety teams under current leadership. The findings have drawn scrutiny from regulators in the UK, Europe, and beyond, amid concerns over privacy, exploitation, and harmful AI content.

The Guardian, Jason Wilson – Hundreds of nonconsensual AI images being created by Grok on X, data shows – 8th January 2026