Independent · Nonprofit · Open access
Deep Lex

Individual case record

James v. Cerebras Systems Inc.

IP & CopyrightStatus: Active

Cause: 17:501 Copyright Infringement

View on CourtListener ↗
Court
N.D. Cal.
Docket
4:25-cv-09361
Filed
30 Oct 2025
AI system
Cerebras-GPT

Individual case record

Overview

SUMMARY Darius H. James, an author and performance artist, filed a class action copyright infringement lawsuit against Cerebras Systems Inc. in the U.S. District Court for the Northern District of California. James alleges that Cerebras downloaded and copied pirated books (including his own copyrighted works) from the Books3 dataset contained within The Pile, and used them to train its Cerebras-GPT family of large language models without authorisation or compensation.

FACTS & CLAIMS On 30 October 2025, the Plaintiff, Darius H. James, filed a class action complaint against Cerebras Systems Inc. James is the author of Negrophobia and That's Blaxploitation!, both of which are registered with the U.S. Copyright Office and included in the Books3 dataset. Cerebras is an AI hardware company headquartered in Sunnyvale, California, that developed the Cerebras-GPT series of LLMs ranging from 111 million to 13 billion parameters. The Complaint relies on Cerebras' own published research paper (attached to the Complaint as Exhibit B) as direct evidence that Cerebras pre-trained these models on The Pile dataset, which includes Books3 - a collection of approximately 196,640 books derived from the Bibliotik private tracker, a pirate library. The Complaint also notes that Books3 was removed from Hugging Face in October 2023 due to reported copyright infringement, but Cerebras had already downloaded and used the dataset before that date.

The Complaint alleges a single cause of action: direct copyright infringement under 17 U.S.C. § 501, claiming Cerebras violated the plaintiff's exclusive rights by copying, storing, processing, and using the copyrighted works to develop and train its models. The case is brought as a class action on behalf of all U.S. copyright holders whose works were included in Cerebras' pre-training data.

The case follows the established shadow library / Books3 litigation pattern seen in cases like Tremblay v. OpenAI and Bartz v. Anthropic. The case also raises the question of whether releasing trained models under an open-source licence (Apache 2.0) constitutes further distribution of infringing derivative works.

The complaint seeks statutory damages, actual damages, restitution of profits, a declaration of wilful infringement, and destruction of all infringing copies under 17 U.S.C. § 503(b).

This summary is provided for educational and research purposes only and does not constitute legal advice. Allegations described are those made in the filings. Please consult the underlying source documents.

Source documents

Documents

Read the complaint and available amendments. For the latest filings, view the source docket ↗.

Loading document records…

Connections

Related content

Models & Systems

Procedural record

Court & docket history

Docket entries from the past three months. For the complete history and latest filings, follow the source docket.

View full docket on CourtListener ↗

Loading docket entries…

Latest filing recorded: 5 Oct 2026.