Elon Musk once called removing child sexual abuse material from X his top priority. Yet new investigations suggest the platform still struggles to keep known abusive content offline. Reviews by The New York Times and the Canadian Center for Child Protection found previously flagged material on X, while Grok generated sexualized images involving children earlier this year.
Grok generated images involving known abuse victims
The Canadian center identified 65 cases where Grok created sexualized or exploitative images involving children before X tightened the chatbot’s restrictions. Some of these children are already known to authorities as abuse victims.
Users were also able to ask Grok to alter photographs of known victims. The chatbot could place clothed children in lingerie or bikinis and create other sexualized scenarios. The activity occurred in December and January, when Grok was generating millions of images after user prompts. Following public criticism, X said it would stop the chatbot from producing such material.
Musk had previously promised that users attempting to generate child abuse material would face consequences similar to those imposed on people sharing real abuse content. That could include account bans.
Separately, The Times ran its own automated search and found more than 75 previously flagged abuse images posted on X between January and June. Some of these posts stayed live long enough to rack up hundreds of views before being taken down. This mirrors a wider trend, as UK child safety reports of explicit deepfakes have already surpassed all of last year’s totals just six months into 2026.
How is X responding to all this?
X maintains it has zero tolerance for this kind of content, including material generated by AI, and says it uses automated filters plus proprietary detection tools. The company reports it has taken action against 780,000 accounts in the past month alone and sent 1.3 million images to the National Center for Missing and Exploited Children this year, leading to 543 arrests.
Still, xAI now faces multiple lawsuits over Grok generating explicit images of minors, and even acknowledged in a May regulatory filing that this risk could affect investors. However, the latest findings show that flagged content can still slip past X’s safeguards.
