.png){kind=logo}
Apple has been hit with a proposed class action lawsuit in federal court in California, on claims that it utilized copyrighted books to train its new model, Apple Intelligence. The plaintiffs, neuroscientists Susana Martinez-Conde and Stephen Macknik, allege that Apple used unlawful so-called shadow libraries of pirated material to train its artificial intelligence system.
According to filings submitted by Reuters, professors at SUNY Downstate Health Sciences University in Brooklyn report that thousands of copied books and other copyrighted materials were included in the AI training datasets provided by Apple. The complaint also cites their own books as examples of the allegedly unauthorized books: Champions of Illusion and Sleights of Mind.
The lawsuit claims unspecified monetary damages and seeks an injunction from the court against Apple's further use of the copyrighted content in its AI systems. Apple has not yet published a response to the allegations.
This is the second copyright lawsuit in recent months to accuse Apple of copyright infringements associated with AI training. A different set of authors has also made a similar complaint in September, claiming that the company had used creatives without obtaining the necessary licenses.
These cases further put Apple in a list of increasing numbers of technology firms, such as OpenAI, Microsoft, and Meta Platforms, that have been sued by authors, journalists, and musicians over allegations of intellectual property abuse in creating artificial intelligence tools.
The complaint highlights the financial aspects of Apple's AI initiatives. It claims that the company's market value increased by over $200 billion following the release of Apple Intelligence, and it was the most profitable day in the history of the company, as the plaintiffs refer to.
The case against Apple highlights the growing tension between authors and technology firms in the context of AI data collection. The rights holders claim that using copyrighted materials to train AI models without a license can be considered a large-scale infringement.
With the growing presence of human-made content in more AI systems, legal results of such cases may determine the future of AI model training and creator rights in the digital era.
Also Read: Judge Rejects Musk’s Bid to Block OpenAI’s For-Profit Shift, But Lawsuit Continues
