Regulation
Meta must say which bosses approved downloading pirated books, Ars Technica reports
Meta blamed its employees for downloading pirated books to train its AI. A judge now requires it to identify who made those decisions, according to Ars Technica.

The essentials
- Claimed by one partyJudge Thomas Hixon has essentially ordered Meta to identify the people who authorized downloading pirated books to train its AI, according to Ars Technica (Ars Technica).
- Claimed by one partyThe books came from websites that share them without permission, such as Anna’s Archive and LibGen, and were downloaded with programs that let users share files with each other (Ars Technica).
- Claimed by one partyThe order affects a class action lawsuit brought by authors, including Sarah Silverman, and another lawsuit brought by the publisher Entrepreneur Media (Ars Technica).
Why it matters
Meta was trying to present the piracy as the work of individual employees. If it has to name supervisors and executives, responsibility could move up the chain of command. That would carry weight in lawsuits over whether training AI on copied books is legal.
The details
According to Ars Technica, Meta argues in a court filing that it first gathered its data “informally” and that no one with authority decided anything. The company says a decision existed only in the sense that some employees “technically” downloaded the files. Those employees testified that their superiors gave them the go-ahead (Ars Technica).
What we don't know
- Which specific executives will be named, or when Meta will have to hand over their names.
- On which points the judge did not side with the plaintiffs: the source says he sided with them “in large part.”
- What effect the order will have on the final outcome of the lawsuits.
Sources
Was this story useful? Yes Not really