Three complaints, one defendant with a new name
Three copyright complaints were filed on 4 November in the US District Court for the Northern District of California against the company formerly known as xAI, which the filings name as SpaceX Super Intelligence Corp., SpaceX Super Intelligence LLC and SpaceX Super Intelligence Holdings LLC, each listed as formerly X.AI.
The plaintiffs are the academic publisher Cognella (case 3:26-cv-13189), Chicken Soup for the Soul (3:26-cv-13188) and Cambronne with four individual authors (3:26-cv-13187). All three are brought by Stris & Maher with Freedman Normand Friedland, and all three run the same three counts.
The count that is not about shadow libraries
Count I is the familiar claim: that the company took the works from pirate libraries. Count II is the one worth reading. It alleges the company bought physical copies, scanned them, ran the images through optical character recognition, then “processed, cleaned, de-duplicated, tokenized, stored, and retained” the digital copies to train and evaluate the Grok family.
Buying the book, the complaints argue, authorises none of that: “Ownership of a physical copy conveys no right to reproduce the copyrighted work embodied in it.” That is a direct run at the holding in Bartz v. Anthropic, where Judge William Alsup found that converting lawfully purchased print books into digital copies for training was fair use. These complaints cite that case for the facts of the scanning programme and then ask a different court to come out the other way.
What the complaints quote
The evidence the plaintiffs lean on is largely the defendant’s own material. They quote the company’s EU AI Act disclosure for Grok 4.7, a public summary published in September, which states that the model was pre-trained on more than 10 trillion tokens, that its crawler drew on “community and document-sharing sites”, and that the company uses the US Trade Representative’s Notorious Markets list “as a signal” when deciding which sites to exclude. The same document says the company concluded commercial licensing agreements with rightsholders for text — but not, the plaintiffs say, with them.
They also quote Elon Musk. In a post from July 2026 cited in all three complaints, he wrote that he had asked the team to preserve rare books in a library and scan them “the hard way vs just cutting off the spine”. The complaints read that as confirmation that the company acquires and scans physical books, and that for ordinary books it used the method Musk described as the alternative.
What they are asking for
Each complaint brings direct infringement on the pirated copies, direct infringement on the scanning, and contributory infringement, and each demands a jury trial. The plaintiffs argue at length that statutory damages must be set by a jury rather than the court, which is the procedural fight that decides how large an AI copyright case can get.
No response has been filed. What to watch is whether the Northern District consolidates these with the existing cases against the company, and whether any judge there is willing to split with Alsup on the scanning question — the point on which the entire print-acquisition strategy of several AI developers now rests.
