xAI sues Grok users over deepfakes, testing a new liability strategy for AI platforms

xAI is suing two Grok users for allegedly creating child sexual abuse material, seeking to recover legal costs from the users rather than relying solely on account suspensions. The strategy is unusual and could reshape how AI platforms handle misuse liability.

What xAI did and why it matters for builders

Elon Musk's xAI has filed lawsuits against two users of its Grok chatbot, alleging they violated X's terms of service by using the tool to create explicit images of children. The company wants those users to cover "all reasonable expenses" xAI incurs defending itself against lawsuits filed by alleged victims, plus damages for reputational harm. Both users, Russell Bloodworth and Terry Wayne Harwood, also face criminal charges related to the same alleged conduct.

This is an aggressive legal move. Most tech companies enforce terms of service through account suspensions and terminations, not by suing users. As law professor John Coyle told POLITICO, "It's not something you really see any other companies doing." xAI has made nearly 74,000 reports of suspected child sexual abuse material to the National Center for Missing and Exploited Children in 2026 alone, resulting in more than 240 arrests, according to court filings. But in these two cases, the company escalated to litigation.

The practical implications for AI platforms

For AI builders, this case tests whether terms of service can be used to shift liability for misuse back to users. If xAI succeeds, other platforms may follow with similar cost-recovery provisions in their terms. That would change the risk calculus for both companies and users.

The lawsuits also highlight the adequacy of technical safeguards. xAI claims it "builds in technological safeguards to prevent bad actors from engaging in illegal, harmful, or abusive conduct." But victims' attorneys argue the company fails to use industry-standard protections. One class action plaintiff's attorney told POLITICO that the volume of Grok-related CSAM reports "indicates to me that the actual problem is with their product, which they should fix to prevent this conduct from happening in the first place."

What remains uncertain

The lawsuits are in early stages. xAI has not explained how it decides which users to sue versus suspend. The company is also defending against at least two other class actions and multiple individual suits over Grok-generated nudification. A San Jose judge recently blocked xAI's bid to name anonymous victims in one case, citing threat of more abuse. The legal strategy may not hold up in court, and as attorney Derek Potts noted, xAI may struggle to recover significant sums from the individual defendants.

For now, builders should watch how these cases unfold. They could set precedent for how AI companies allocate responsibility between their own safeguards and user behavior.