Jump to content

Doe v X.AI Corp: Difference between revisions

From AI Law Wiki
Migration export
Update Doe v X.AI Corp with amended complaint adding Stability AI
 
Line 12: Line 12:
* '''X.AI Corp.'''<ref name="prokopiev" />
* '''X.AI Corp.'''<ref name="prokopiev" />
* '''X.AI LLC'''<ref name="prokopiev" />
* '''X.AI LLC'''<ref name="prokopiev" />
* '''Stability AI''' — added as a defendant in the July 8, 2026 amended complaint.<ref name="lieff-amended">[https://www.lieffcabraser.com/2026/07/deepfake-victims-bolster-class-action-against-xai-add-stability-ai/ Lieff Cabraser, "Deepfake Victims Bolster Class Action Against xAI, Add Stability AI" (July 8, 2026).]</ref>


Counsel for plaintiffs: Lieff Cabraser Heimann & Bernstein.<ref name="lieffcabraser" />
Counsel for plaintiffs: Lieff Cabraser Heimann & Bernstein.<ref name="lieffcabraser" />
Line 60: Line 61:


* '''[[News xAI-v-Bonta-California-AI-Transparency-Ruling-2026|xAI v. Bonta]]''' — xAI's challenge to California AI Transparency Law (separate proceeding)
* '''[[News xAI-v-Bonta-California-AI-Transparency-Ruling-2026|xAI v. Bonta]]''' — xAI's challenge to California AI Transparency Law (separate proceeding)
== July 2026 amended complaint ==
On July 8, 2026, plaintiffs filed an amended complaint adding two new minor plaintiffs and naming Stability AI as a defendant.<ref name="lieff-amended" /> The amended complaint alleges that xAI failed to provide user-identifying information in CyberTipline reports and that Stability AI models enabled downstream nudification applications used to generate CSAM.<ref name="lieff-amended" /> Ars Technica reported that the amendment included allegations that one perpetrator used Grok to create roughly 7,000 sexually explicit images from a photograph of an 11-year-old child.<ref name="ars-amended">[https://arstechnica.com/tech-policy/2026/07/lawsuit-grok-user-made-7k-child-sex-images-xai-only-reported-one-gang-rape-prompt/ Ars Technica, "Lawsuit: Man used Grok to make 7K sex images of stepdaughter, then shot himself" (July 8, 2026).]</ref>


== References ==
== References ==

Latest revision as of 01:33, 12 July 2026

Doe 1 v. X.AI Corp. (Case No. 5:26-cv-02246, N.D. Cal.) is a class action lawsuit filed on March 16, 2026, alleging that xAI's Grok AI model generated child sexual abuse material (CSAM) deepfakes using real photographs of minors.[1][2]

Parties

[edit]

Plaintiffs

[edit]
  • Jane Doe 1 — minor whose real image was used to generate CSAM via Grok
  • Jane Doe 2 — minor victim
  • Jane Doe 3 — minor victim
  • Putative class: All U.S. persons whose real minor images were altered by Grok into sexualized images or videos[1]

Defendants

[edit]
  • X.AI Corp.[2]
  • X.AI LLC[2]
  • Stability AI — added as a defendant in the July 8, 2026 amended complaint.[3]

Counsel for plaintiffs: Lieff Cabraser Heimann & Bernstein.[1]

Court

[edit]
  • Court: U.S. District Court for the Northern District of California (San Jose Division)[2]
  • Judge: Not yet assigned; plaintiffs filed motion to relate case to prior matter for potential same-judge assignment (Docket No. 3)[2]

Claims

[edit]
  • Masha's Law (18 U.S.C. § 2255) — civil remedy for child sexual exploitation victims[1]
  • Trafficking Victims Protection Act[1]
  • California state law claims — negligence, products liability[2]

Relief sought: Damages, punitive damages, and injunctive relief.[1]

Factual Background

[edit]

The complaint alleges that Grok's "Spicy Mode" — a feature restricted to paid subscribers — enabled users to upload real photographs of minors and generate photorealistic nude and sexually explicit images and videos.[1] Reports indicate Grok generated approximately 3 million sexualized images and 23,000 apparent child depictions in late 2025–early 2026 before xAI restricted the feature after public exposure.[1][4]

The California Attorney General launched an investigation into xAI/Grok on January 14, 2026, citing nonconsensual deepfake intimate images of women, girls, and children.[5]

One perpetrator who used Grok to generate CSAM was arrested; CSAM was reported to NCMEC.[1]

Procedural History

[edit]
Date Docket No. Event
March 16, 2026 1 Complaint filed (44 pages; filing fee $405; nature of suit 360 P.I.: Other Personal Injury)[2]
March 16, 2026 3 Motion to relate case to prior matter filed[2]

No answer, motion to dismiss, or other responsive filing by xAI has been reported as of April 2026.[2]

Significance

[edit]

This is one of the first class action lawsuits alleging that an AI model directly generated CSAM from real minors' images, raising novel questions about:

  • AI company liability for model outputs that violate federal child exploitation laws
  • Adequacy of safety guardrails in commercial AI products
  • Scope of civil remedies under Masha's Law and the TVPA for AI-generated CSAM
  • Product liability theories applied to generative AI
[edit]
  • xAI v. Bonta — xAI's challenge to California AI Transparency Law (separate proceeding)


July 2026 amended complaint

[edit]

On July 8, 2026, plaintiffs filed an amended complaint adding two new minor plaintiffs and naming Stability AI as a defendant.[3] The amended complaint alleges that xAI failed to provide user-identifying information in CyberTipline reports and that Stability AI models enabled downstream nudification applications used to generate CSAM.[3] Ars Technica reported that the amendment included allegations that one perpetrator used Grok to create roughly 7,000 sexually explicit images from a photograph of an 11-year-old child.[6]

References

[edit]