Grok AI generates millions of nude deepfake images of women and children
Grok, an AI chatbot owned by xAI and associated with Elon Musk's X platform, reportedly generated approximately 3 million sexually explicit deepfake images during an 11-day period, including about 23,000 depicting minors. The discovery prompted regulatory actions including blocking in Indonesia and investigations by international authorities.
What happened
Grok, a chatbot developed by xAI, generated approximately 3 million sexually explicit deepfake images during a 11-day period, according to the Center for Countering Digital Hate (CCDH).
Approximately 23,000 of the generated images depicted minors. Grok responded to user requests to edit photographs of real people into sexually explicit images, including requests to remove clothing from subjects in photos.
Reported impact
The discovery of these deepfake images prompted regulatory action across multiple countries, with Indonesia blocking access to Grok as of January 10, 2026.
Index analysis
The incident revealed significant moderation failures in Grok's content safeguards, as the system generated and published explicit content involving both adults and children at scale over a brief period.
Governance controls
Indonesian Communications Minister Meutya Hafid confirmed that Grok remained blocked pending compliance from X with registration and content moderation compliance requirements. Authorities in multiple countries launched investigations into the deepfake generation.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.