Baltimore has filed what CNBC describes as the first U.S. city-level lawsuit against Elon Musk's xAI, X Corp., and SpaceX, alleging that the Grok AI chatbot was knowingly deployed as a tool for generating non-consensual intimate imagery — including material depicting minors.

The complaint, filed in Maryland court and represented by law firm DiCello Levitt, claims Grok generated between 1.8 million and 3 million sexualized images in a nine-day window between December 29, 2025, and January 8, 2026, with around 23,000 estimated to depict children.

The surge in output was partly attributed to Elon Musk himself. After he responded "Perfect" to a Grok-generated bikini image of himself posted on X, daily output reportedly jumped from roughly 300,000 images in the nine preceding days to nearly 600,000 per day.

"These deepfakes, especially those depicting minors, have traumatic, lifelong consequences for victims," Baltimore Mayor Brandon M. Scott said in a statement.

The lawsuit alleges the companies violated local consumer protection laws by designing and deploying Grok while publicly claiming such content was prohibited. Baltimore is seeking civil penalties, restitution for affected residents, and injunctions to halt the alleged conduct.

Legal analysts say the case is significant precisely because it arrives at the city level, in the absence of federal AI legislation. Whether courts classify Grok as an "active creator" of harmful content — rather than a passive tool — could determine how AI liability law develops in the U.S. going forward.

The lawsuit joins ongoing investigations by regulators in the EU, France, the UK, Australia, Ireland, and multiple U.S. states.