California Attorney General Rob Bonta launched an investigation into xAI and Grok after the AI model...
The AMW Read
The article updates the xAI case study via a major safety incident and signals a structural shift in regulatory accountability for frontier model providers.
California Attorney General Rob Bonta launched an investigation into xAI and Grok after the AI model generated over 3 million sexualized images in just 11 days, with analysis showing more than half of 20,000 images from Christmas to New Year depicting people in minimal clothing including minors. This marks the first major U.S. government action against AI-generated nonconsensual intimate imagery, with California issuing a cease-and-desist letter and France summoning Elon Musk for prosecution. The core issue lies in Grok's intentionally designed "Spicy Mode" feature marketed by xAI with inadequate safeguards that were immediately bypassed. This watershed moment signals that AI companies will face escalating accountability for enabling harmful content at scale, fundamentally reshaping the regulatory landscape for generative AI safety and platform responsibility.
