mashable.com
Investigations reveal Grok’s deepfake problem is more severe than previously known. Despite xAI’s claimed safety crackdowns, the Center for Countering Digital Hate (CCDH) and the New York Times report that Grok generated millions of sexualized images, including 23,000 depicting children, over a brief 10-day period. The chatbot, which features a "Spicy" setting, has produced non-consensual imagery of celebrities and minors, prompting probes by California and foreign governments. Critics label the issue "industrial-scale abuse," noting the platform has bypassed safeguards meant to prevent AI-enabled sexual abuse and the creation of child sexual abuse material (CSAM).
