Media & Culture

xAI sues Grok user for creating illegal deepfakes after promoting uncensored AI

73,604 CSAM reports filed, yet Grok's 'spicy mode' let users undress anyone.

Deep Dive

Elon Musk's xAI has filed a lawsuit against South Carolina user Terry Wayne Harwood for allegedly using its Grok chatbot to create child sexual abuse material (CSAM) and non-consensual sexual imagery. Filed in federal court in Texas, the suit cites repeated violations of xAI's Terms of Service and Acceptable Use Policy, which explicitly ban 'undressing or nudifying real persons' and depicting likenesses in a pornographic manner. xAI says it reported 73,604 CSAM instances to the National Center for Missing and Exploited Children (NCMEC) in 2026, contributing to 244 arrests. The company argues Grok is a 'neutral tool,' putting full responsibility on users.

However, the lawsuit contradicts Musk's earlier promotion of Grok as an uncensored chatbot. In 2025, Grok offered a 'spicy mode' that allowed users to easily generate non-consensual sexual images of real people, including women and children. Gizmodo tests confirmed it was easier to undress women than men. UK officials threatened X with fines or bans in January. Musk, who previously posted sexualized Grok videos, has since deleted some. Reddit forums and FTC complaints show users frustrated with new guardrails. The suit appears to be an attempt to distance xAI from a culture Musk himself cultivated.

Key Points
  • xAI sues Terry Wayne Harwood in Texas for creating CSAM and deepfakes using Grok, citing violation of terms banning 'undressing' real persons.
  • xAI reported 73,604 CSAM cases to NCMEC in 2026, leading to 244 arrests, but Grok's 'spicy mode' previously allowed easy generation of sexual deepfakes.
  • Musk's earlier promotion of Grok as uncensored, including a 2025 video of sexualized anime girls, contradicts the lawsuit's claim of strong safeguards.

Why It Matters

The lawsuit highlights the tension between AI uncensorship as a marketing tactic and the legal reality of CSAM, risking reputational damage and regulatory action.

📬 Get the top 10 AI stories daily