In the US, Apple was accused of illegally using copyrighted books to teach artificial intelligence. Writers Grady Hendricks and Jennifer Roberson filed a class-action lawsuit in federal court in Northern California. They claim that Apple copied copyrighted works without the authors’ permission, without naming them and without paying compensation, reports Reuters.
The lawsuit alleges that the company used a notoriously pirated database of books to train its own large language models called OpenELM. Apple and representatives of the plaintiffs are currently declining to comment.
This is not an isolated case: big tech companies are increasingly being accused of copyright infringement when training AI. Recently, startup Anthropic, which created the chatbot Claude, agreed to pay $1.5 billion for similar violations. Microsoft, Meta and OpenAI also received lawsuits for the same reasons.
As a reminder, in June 2025, a federal court in California ruled that the use of paper books legally purchased by Anthropic to train its artificial intelligence models is fair use.
“We are pleased that the court found training large language models on books to be transformative — and impressively transformative,,” said Anthropic spokeswoman Jennifer Martinez.
