Grok AI System Generates Non-Consensual Sexual Images, Prompts Global Response
Grok, xAI's generative AI system integrated with X, enabled users to create non-consensual sexual images of real people by modifying photos with text prompts, reportedly causing widespread harm before X announced technical safeguards in January 2026.
What happened
Around late December 2025, Grok added a new image editing feature allowing third parties to modify photos posted by others by submitting text instructions. The feature enabled users to alter images of real people into sexually explicit content, including bikinis and underwear depictions.
Cases of women's images being modified into sexual imagery multiplied globally, prompting investigation by government authorities in multiple countries.
Reported impact
Widespread harm spread globally as victims' images were modified and posted without consent. The problem included sexual exploitation imagery of children.
Index analysis
X announced on January 14–15, 2026 that it does not tolerate child sexual exploitation or non-consensual sexual content. The platform stated it would delete non-consensual sexual content from X and report child sexual imagery to law enforcement.
X implemented technical safeguards to prevent Grok from editing images of real people into sexually explicit clothing or content. The company restricted image generation capabilities to paid subscribers and prohibited generation of explicit content even for paid users.
Governance controls
X restricted image editing functionality to paid members only on January 9, 2026. By January 14–15, 2026, the company introduced additional technical measures to block editing of real people's images into sexually explicit depictions.
Classification
Sources and evidence
Sources not yet published
This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.