Meta is arguing in a federal lawsuit that uploading pirated books to other BitTorrent users during its LLM training data acquisition qualifies as fair use. The company claims the uploading was an unavoidable technical consequence of using BitTorrent to download datasets from shadow libraries like Anna's Archive, and therefore falls under the same fair use purpose as the training itself. Authors' lawyers contest this defense as improperly raised after discovery deadlines. Notably, deposed authors admitted they are unaware of any Meta model output replicating their work, which Meta uses to argue there is no market harm. Judge Chhabria must now decide whether to allow this 'fair use by technical necessity' defense, a ruling that could have broad implications for AI copyright cases involving shadow libraries.
Table of contents
Meta’s Bittersweet VictorySeeding Pirated Books is Fair UseAuthors and Meta Disagree over Fair Use TimingAuthors Admit No Harm, No Infringing Output‘U.S. AI Leadership at Stake’Sort: