xAI Sues User for Generating CSAM with Grok, Seeks to Shield Itself from Liability

xAI has filed a lawsuit against Terry Wayne Harwood, a South Carolina man arrested for possession and distribution of child sexual abuse materials (CSAM) allegedly generated using its Grok chatbot. The complaint, reported by Ars Technica, marks xAI’s first legal action against a user for creating illegal content with its AI tool and comes amid a proposed class action claiming xAI failed to prevent Grok from generating CSAM and then obstructed law enforcement.

The lawsuit alleges that Harwood used two accounts from December 2025 to February 2026 to generate non-consensual explicit images of multiple victims, including a minor as young as 10. While xAI acknowledges that Grok sometimes refused prompts—such as a request to “remove all her clothing”—the company did not disclose specific methods Harwood used to bypass safeguards.

In its legal filing, xAI argues that Grok is “a neutral tool, subject to user control,” and that Harwood alone is liable for violating its terms of service, which prohibit undressing or sexualizing real people. The company is seeking a court ruling that users, not xAI, bear responsibility for outputs, and that an indemnity clause in its terms holds users liable for any harm. This defense strategy aims to undermine the proposed class action, which involves thousands of potential victims, and to reduce xAI’s exposure to damages.

Critics point out that the Copyright Office does not consider AI outputs as human creations, complicating the argument that users are the sole authors—and thus solely liable—of generated content. Meanwhile, the lawsuit follows reports that xAI filed 90% of its CSAM reports to NCMEC without user identifying information, hindering investigations.